Sample records for differential imaging features

  1. Registration of image feature points using differential evolution

    ZHANG Hao; HUANG Zhan-hua; YU Dao-ying


    This paper introduces a robust global nonlinear optimizer-differential evolution(DE),which is a simple evolution algorithm to search for an optimal transformation that makes the best alignment of two sets of feature points.To map the problem of matching into the framework of DE,the objective function is proportional to the registration error which is measured by Hausdorff distance,while the parameters of transformation are encoded in floating-point as the functional variables.Three termination criteria are proposed for DE.A simulation of 2-dimensional point sets and a similarity transformation are presented to compare the robustness and convergence properties of DE with genetic algorithm's (GA).And the registration of an object and its contour model have been demonstrated by using of DE to natural images.

  2. Imaging Characteristics of Pathologically Proven Thymic Hyperplasia: Identifying Features That Can Differentiate True From Lymphoid Hyperplasia

    Araki, Tetsuro; Sholl, Lynette M.; Gerbaudo, Victor H.; Hatabu, Hiroto; Nishino, Mizuki


    OBJECTIVE The purpose of this article is to investigate the imaging characteristics of pathologically proven thymic hyperplasia and to identify features that can differentiate true hyperplasia from lymphoid hyperplasia. MATERIALS AND METHODS Thirty-one patients (nine men and 22 women; age range, 20–68 years) with pathologically confirmed thymic hyperplasia (18 true and 13 lymphoid) who underwent preoperative CT (n = 27), PET/CT (n = 5), or MRI (n = 6) were studied. The length and thickness of each thymic lobe and the transverse and anterior-posterior diameters and attenuation of the thymus were measured on CT. Thymic morphologic features and heterogeneity on CT and chemical shift on MRI were evaluated. Maximum standardized uptake values were measured on PET. Imaging features between true and lymphoid hyperplasia were compared. RESULTS No significant differences were observed between true and lymphoid hyperplasia in terms of thymic length, thickness, diameters, morphologic features, and other qualitative features (p > 0.16). The length, thickness, and diameters of thymic hyperplasia were significantly larger than the mean values of normal glands in the corresponding age group (p hyperplasia was significantly higher than that of true hyperplasia among 15 patients with contrast-enhanced CT (median, 47.9 vs 31.4 HU; Wilcoxon p = 0.03). The receiver operating characteristic analysis yielded greater than 41.2 HU as the optimal threshold for differentiating lymphoid hyperplasia from true hyperplasia, with 83% sensitivity and 89% specificity. A decrease of signal intensity on opposed-phase images was present in all four cases with in- and opposed-phase imaging. The mean maximum standardized uptake value was 2.66. CONCLUSION CT attenuation of the thymus was significantly higher in lymphoid hyperplasia than in true hyperplasia, with an optimal threshold of greater than 41.2 HU in this cohort of patients with pathologically confirmed thymic hyperplasia. PMID:24555583

  3. Sclerosing cholangitis: Clinicopathologic features, imaging spectrum, and systemic approach to differential diagnosis

    Seo, Ni Eun [Dept. of Radiology, Research Institute of Radiological Science, Severance Hospital, Yonsei University College of Medicine, Seoul (Korea, Republic of); Kim, So Yeon; Lee, Seung Soo; Byun, Jae Ho; Kim, Hyoung Jung; Kim, Jin Hee; Lee, Moon Gyu [Dept. of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul (Korea, Republic of)


    Sclerosing cholangitis is a spectrum of chronic progressive cholestatic liver disease characterized by inflammation, fibrosis, and stricture of the bile ducts, which can be classified as primary and secondary sclerosing cholangitis. Primary sclerosing cholangitis is a chronic progressive liver disease of unknown cause. On the other hand, secondary sclerosing cholangitis has identifiable causes that include immunoglobulin G4-related sclerosing disease, recurrent pyogenic cholangitis, ischemic cholangitis, acquired immunodeficiency syndrome-related cholangitis, and eosinophilic cholangitis. In this review, we suggest a systemic approach to the differential diagnosis of sclerosing cholangitis based on the clinical and laboratory findings, as well as the typical imaging features on computed tomography and magnetic resonance (MR) imaging with MR cholangiography. Familiarity with various etiologies of sclerosing cholangitis and awareness of their typical clinical and imaging findings are essential for an accurate diagnosis and appropriate management.

  4. Neuroendocrine differentiated breast carcinoma: imaging features correlated with clinical and histopathological findings

    Guenhan-Bilgen, Isil; Ustuen, Esin Emin; Memis, Aysenur [Department of Radiology, Ege University Hospital, Bornova, 35100 Izmir (Turkey); Zekioglu, Osman; Erhan, Yildiz [Department of Pathology, Ege University Hospital, Bornova, 35100 Izmir (Turkey)


    The aim of this study was to describe the imaging features of neuroendocrine differentiated breast carcinoma (NEDBC) and to correlate the radiological findings with the clinical and histopathological findings. A retrospective review of the mammograms of 1845 histopathologically proven breast cancer cases revealed five NEDBC. The clinical, imaging, and histopathological findings were analyzed. On mammography, a high-density mass was seen in all patients. The shape of the mass was round in 4 and irregular in 1 patient. The margins were spiculated in 2, indistinct in 1, microlobulated in 1, and partially obscured in 1 patient. On sonography, 4 patients had homogeneously hypoechoic masses with normal sound transmission. In 1 patient the mass was heterogeneously hypoechoic with mild posterior acoustic enhancement. The margins were microlobulated in 2, irregular in 2, and well-circumscribed in 1 patient. Neuroendocrine differentiated breast carcinoma should be included in the differential diagnosis of mammographically dense, round masses with predominantly spiculated or lobulated margins. Sonographically, they mostly present as irregular or microlobulated, homogeneously hypoechoic masses with normal sound transmission. (orig.)

  5. Differential diagnosis of dumbbell lesions associated with spinal neural foraminal widening: Imaging features

    Kivrak, Ali Sami [Selcuk University, Meram Medical Faculty, Department of Radiology, 42080 Konya (Turkey)], E-mail:; Koc, Osman; Emlik, Dilek; Kiresi, Demet; Odev, Kemal [Selcuk University, Meram Medical Faculty, Department of Radiology, 42080 Konya (Turkey); Kalkan, Erdal [Selcuk University, Meram Medical Faculty, Department of Neurosurgery, Konya (Turkey)


    Computed tomography (CT) and magnetic resonance imaging (MRI) reliably demonstrate typical features of schwannomas or neurofibromas in the vast majority of dumbbell lesions responsible for neural foraminal widening. However, a large variety of unusual lesions which are causes of neural foraminal widening can also be encountered in the spinal neural foramen. Radiologic findings can be helpful in differential diagnosis of lesions of spinal neural foramen including neoplastic lesions such as benign/malign peripheral nerve sheath tumors (PNSTs), solitary bone plasmacytoma (SBP), chondroid chordoma, superior sulcus tumor, metastasis and non-neoplastic lesions such as infectious process (tuberculosis, hydatid cyst), aneurysmal bone cyst (ABC), synovial cyst, traumatic pseudomeningocele, arachnoid cyst, vertebral artery tortuosity. In this article, we discuss CT and MRI findings of dumbbell lesions which are causes of neural foraminal widening.

  6. Differentiating benign from malignant bone tumors using fluid-fluid level features on magnetic resonance imaging

    Yu, Hong; Cui, Jian Ling; Cui, Sheng Jie; Sun, Ying Cal; Cui, Feng Zhen [Dept. of Radiology, The Third Hospital of Hebei Medical University, Hebei Province Biomechanical Key Laborary of Orthopedics, Shijiazhuang, Hebei (China)


    To analyze different fluid-fluid level features between benign and malignant bone tumors on magnetic resonance imaging (MRI). This study was approved by the hospital ethics committee. We retrospectively analyzed 47 patients diagnosed with benign (n = 29) or malignant (n = 18) bone tumors demonstrated by biopsy/surgical resection and who showed the intratumoral fluid-fluid level on pre-surgical MRI. The maximum length of the largest fluid-fluid level and the ratio of the maximum length of the largest fluid-fluid level to the maximum length of a bone tumor in the sagittal plane were investigated for use in distinguishing benign from malignant tumors using the Mann-Whitney U-test and a receiver operating characteristic (ROC) analysis. Fluid-fluid level was categorized by quantity (multiple vs. single fluid-fluid level) and by T1-weighted image signal pattern (high/low, low/high, and undifferentiated), and the findings were compared between the benign and malignant groups using the chi2 test. The ratio of the maximum length of the largest fluid-fluid level to the maximum length of bone tumors in the sagittal plane that allowed statistically significant differentiation between benign and malignant bone tumors had an area under the ROC curve of 0.758 (95% confidence interval, 0.616-0.899). A cutoff value of 41.5% (higher value suggests a benign tumor) had sensitivity of 73% and specificity of 83%. The ratio of the maximum length of the largest fluid-fluid level to the maximum length of a bone tumor in the sagittal plane may be useful to differentiate benign from malignant bone tumors.

  7. CT and magnetic resonance imaging features of middle ear adenoma of neuroendocrine differentiation: A case report

    Lee, Sang Kwon; Choe, Mi Sun [Keimyung University School of Medicine, Daegu (Korea, Republic of)


    Middle ear adenoma is a rare benign epithelial tumor. We report the CT and magnetic resonance imaging findings of a case of middle ear adenoma of neuroendocrine differentiation in a 36-year-old man. On high-resolution CT, the mass was found to fill the middle ear, in which the ossicles were embedded, but not destroyed, with outward bulging of the intact tympanic membrane. On MRI, the mass, which was intensely enhanced on 3-dimensional (3D) gadolinium (Gd)-enhanced spoiled gradient-recalled (SPGR) sequence, involved the middle ear, aditus ad antrum and a portion of mastoid antrum. Histological and immunohistochemical findings of the specimen obtained by surgical excisions were consistent with middle ear adenoma of neuroendocrine differentiation. Middle ear adenoma of neuroendocrine differentiation should be included in the differential diagnosis of an intensely enhancing mass filling the middle ear/mastoid antrum without ossicular destructions. The extent of the mass can be excellently assessed with 3D Gd-enhanced SPGR sequence.

  8. Morphologic MRI features, diffusion tensor imaging and radiation dosimetric analysis to differentiate pseudo-progression from early tumor progression.

    Agarwal, Ajay; Kumar, Sanath; Narang, Jayant; Schultz, Lonni; Mikkelsen, Tom; Wang, Sumei; Siddiqui, Sarmad; Poptani, Harish; Jain, Rajan


    Pseudo-progression (PsP) refers to the paradoxical increase of contrast enhancement within 12 weeks of chemo-radiation therapy in gliomas attributable to treatment effects rather than early tumor progression (ETP). This study was performed to evaluate the utility of morphologic imaging features, diffusion tensor imaging (DTI) and radiation dosimetric analysis of magnetic resonance imaging (MRI) changes in differentiating PsP from ETP. Serial MRI examinations of 163 patients treated for high-grade glioma were reviewed. 46 patients showed a recurrent or progressive enhancing lesion within 12 weeks of radiotherapy. We used an in-house modified scoring system based on 20 different morphologic features (modified VASARI features) to assess the MRI studies. DTI analyses were performed in 24 patients. MRI changes were defined as recurrent volume (Vrec) and registered with pretreatment computed tomography dataset, and the actual dose received by the Vrec during treatment was calculated using dose-volume histograms. Bidimensional product of T2-FLAIR signal abnormality and enhancing component was larger in the ETP group. DTI metrics revealed no significant difference between the two groups. There was no statistically significant difference in the location of Vrec between PsP and ETP groups. Morphologic MRI features and DTI have a limited role in differentiating between PsP and ETP. The larger sizes of the T2-FLAIR signal abnormality and the enhancing component of the lesion favor ETP. There was no correlation between the pattern of MRI changes and radiation dose distribution between PsP and ETP groups.

  9. Abdominal tuberculosis: Imaging features

    Pereira, Jose M. [Department of Radiology, Hospital de S. Joao, Porto (Portugal)]. E-mail:; Madureira, Antonio J. [Department of Radiology, Hospital de S. Joao, Porto (Portugal); Vieira, Alberto [Department of Radiology, Hospital de S. Joao, Porto (Portugal); Ramos, Isabel [Department of Radiology, Hospital de S. Joao, Porto (Portugal)


    Radiological findings of abdominal tuberculosis can mimic those of many different diseases. A high level of suspicion is required, especially in high-risk population. In this article, we will describe barium studies, ultrasound (US) and computed tomography (CT) findings of abdominal tuberculosis (TB), with emphasis in the latest. We will illustrate CT findings that can help in the diagnosis of abdominal tuberculosis and describe imaging features that differentiate it from other inflammatory and neoplastic diseases, particularly lymphoma and Crohn's disease. As tuberculosis can affect any organ in the abdomen, emphasis is placed to ileocecal involvement, lymphadenopathy, peritonitis and solid organ disease (liver, spleen and pancreas). A positive culture or hystologic analysis of biopsy is still required in many patients for definitive diagnosis. Learning objectives:1.To review the relevant pathophysiology of abdominal tuberculosis. 2.Illustrate CT findings that can help in the diagnosis.

  10. Tongue base schwannoma: differential diagnosis and imaging features with a case presentation


    Schwannomas are slow growing, encapsulated neoplasms that arise from the nerve sheath. A vast majority of these benign neoplasms occur in the head and neck region, most commonly involving the 8th cranial nerve. Schwannomas arising from the base of tongue are very rare and, thus, can easily escape the list of differential diagnosis for a posterior tongue mass. A systematic approach is recommended for diagnosis of a posterior tongue mass, with neoplastic, infectious, and congenital categories. ...

  11. Multispectral Image Feature Points

    Cristhian Aguilera


    Full Text Available This paper presents a novel feature point descriptor for the multispectral image case: Far-Infrared and Visible Spectrum images. It allows matching interest points on images of the same scene but acquired in different spectral bands. Initially, points of interest are detected on both images through a SIFT-like based scale space representation. Then, these points are characterized using an Edge Oriented Histogram (EOH descriptor. Finally, points of interest from multispectral images are matched by finding nearest couples using the information from the descriptor. The provided experimental results and comparisons with similar methods show both the validity of the proposed approach as well as the improvements it offers with respect to the current state-of-the-art.

  12. [Intracranial chondroma arising from the skull base: two case reports featuring the image findings for differential diagnosis].

    Higashida, Tetsuhiro; Sakata, Katsumi; Kanno, Hiroshi; Tanabe, Yutaka; Kawasaki, Takashi; Yamamoto, Isao


    We reported two cases of intracranial skull base chondroma and discussed the differential diagnosis and the treatment strategies. The first case was a 39-year-old male who presented with left exophtalmos, visual loss and oculomotor disturbance. MRI showed a huge tumor occupying the bilateral cavernous sinus. Partial removal of the tumor was performed through the left orbitozygomatic subtemporal approach. The second case was a 54-year-old male who presented with left hemiparesis. MRI showed a brain stem infarction with a huge tumor located at the right middle fossa. Partial removal was performed through the right orbitozygomatic subtemporal approach. In these two cases, the histopathological diagnosis of the tumors was benign chondroma and the size of residual tumors have not changed for one year without any additional therapy. Although preoperative definite diagnosis for skull base chondromas is difficult, strategies for diagnosis and treatment without any complication are essential. In our cases, chondromas showed low uptake in PET images, which might be useful for differentiation between chondromas and chordomas. The current popular surgical approach for parasellar tumors is transcranial such as the orbitozygomatic subtemporal approach. In surgical removal of skull base chondromas, it is advisable to try to confirm the diagnosis preoperatively with characteristic image findings and to consider the best approach in each case to decompress the involved nerves without any complications.

  13. Tomographic imaging of the spleen: the role of morphological and metabolic features in differentiating benign from malignant diseases.

    Mainenti, Pier Paolo; Iodice, Delfina; Cozzolino, Immacolata; Segreto, Sabrina; Capece, Sergio; Sica, Giacomo; Magliulo, Mario; Ciancia, Giuseppe; Pace, Leonardo; Salvatore, Marco


    To evaluate the tomographic features in differentiating benign from malignant splenic diseases, 54 patients with a cytohistological examination and a contrast-enhanced multidetector computed tomography (ce-MDCT) and/or positron emission tomography/computed tomography (PET/CT) were retrospectively selected. Significant associations were observed between ce-MDCT Pattern 3 (focal hyperdense lesion) and Pattern 4 (infarcts/cysts) as well as PET/CT Pattern 3 (focal photopenia/diffuse uptake

  14. Correspondence Differential Ghost Imaging

    Li, Ming-Fei; Luo, Kai-Hong; Wu, Ling-An; Fan, Heng


    Experimental data with digital masks and a theoretical analysis are presented for a nonlocal imaging scheme that we name correspondence differential ghost imaging (CDGI). It is shown that by conditional averaging of the information from the reference detector but with the negative signals inverted, the quality of the reconstructed images is, in general, superior to all other ghost imaging (GI) methods to date. The advantages of both differential GI and correspondence GI are combined, plus less data and shorter computation time are required to obtain equivalent quality images under the same conditions. This CDGI method offers a general approach applicable to all GI techniques, especially when objects with continuous gray tones are involved.

  15. Featured Image: Interacting Galaxies

    Kohler, Susanna


    This beautiful image shows two galaxies, IC 2163 and NGC 2207, as they undergo a grazing collision 114 million light-years away. The image is composite, constructed from Hubble (blue), Spitzer (green), and ALMA (red) data. In a recent study, Debra Elmegreen (Vassar College) and collaborators used this ALMA data to trace the individual molecular clouds in the two interacting galaxies, identifying a total of over 200 clouds that each contain a mass of over a million solar masses. These clouds represent roughly half the molecular gas in the two galaxies total. Elmegreen and collaborators track the properties of these clouds and their relation to star-forming regions observed with Hubble. For more information about their observations, check out the paper linked below.A closer look at the ALMA observations for these galaxies, with the different emission regions labeled. Most of the molecular gas emission comes from the eyelids of IC 2163, and the nuclear ring and Feature i in NGC 2207. [Elmegreen et al. 2017]CitationDebra Meloy Elmegreen et al 2017 ApJ 841 43. doi:10.3847/1538-4357/aa6ba5

  16. Localized scleroderma: imaging features

    Liu, P. (Dept. of Diagnostic Imaging, Hospital for Sick Children, Toronto, ON (Canada)); Uziel, Y. (Div. of Rheumatology, Hospital for Sick Children, Toronto, ON (Canada)); Chuang, S. (Dept. of Diagnostic Imaging, Hospital for Sick Children, Toronto, ON (Canada)); Silverman, E. (Div. of Rheumatology, Hospital for Sick Children, Toronto, ON (Canada)); Krafchik, B. (Div. of Dermatology, Dept. of Pediatrics, Hospital for Sick Children, Toronto, ON (Canada)); Laxer, R. (Div. of Rheumatology, Hospital for Sick Children, Toronto, ON (Canada))


    Localized scleroderma is distinct from the diffuse form of scleroderma and does not show Raynaud's phenomenon and visceral involvement. The imaging features in 23 patients ranging from 2 to 17 years of age (mean 11.1 years) were reviewed. Leg length discrepancy and muscle atrophy were the most common findings (five patients), with two patients also showing modelling deformity of the fibula. One patient with lower extremity involvement showed abnormal bone marrow signals on MR. Disabling joint contracture requiring orthopedic intervention was noted in one patient. In two patients with ''en coup de sabre'' facial deformity, CT and MR scans revealed intracranial calcifications and white matter abnormality in the ipsilateral frontal lobes, with one also showing migrational abnormality. In a third patient, CT revealed white matter abnormality in the ipsilateral parietal lobe. In one patient with progressive facial hemiatrophy, CT and MR scans showed the underlying hypoplastic left maxillary antrum and cheek. Imaging studies of areas of clinical concern revealed positive findings in half our patients. (orig.)

  17. Imaging features of gray-scale and contrast-enhanced color doppler US for the differentiation of transient renal arterial ischemia and arterial infarction

    Park, Byung Kwan; Kim Seung Hyup [Seoul National University Hospital, Seoul (Korea, Republic of); Moon, Min Hwan; Jung, Sung Il [Sungkyunkwan University School of Medicine, Seoul (Korea, Republic of)


    To characterize the imaging features on gray-scale and contrast-enhanced color Doppler US images which differentiate renal ischemia from renal infarction. The segmental renal arteries of eight healthy rabbits were surgically ligated. In four of these rabbits, the ligated renal artery was released 60 minutes after arterial occlusion to cause transient ischemia. In the remaining four rabbits, the arterial ligation was retained to cause a permanent infarction. The gray-scale and contrast-enhanced color Doppler US imaging features of the involved renal parenchymal of both ischemia and infarction groups were compared with respect to the presence or absence of parenchymal swelling, echogenicity changes, tissue loss and perfusion defects. Parenchyma swelling, echogenic changes, tissue loss and perfusion defects were found to be more extensive in the infarction than the ischemia group. The hyperechoic areas reperfused with blood flow recovered normal echogenicity and perfusion, whereas the hyperechoic areas without reperfusion became renal infarcts. Gray-scale and contrast-enhanced color Doppler US showed that the hyperechoic areas with reperfusion may reverse to normal parenchyma and allow the differentiation of renal ischemia from renal infarction.

  18. Imaging features of ciliated hepatic foregut cyst

    Song-Hua Fang; Dan-Jun Dong; Shi-Zheng Zhang


    Ciliated hepatic foregut cyst (CHFC) is a very rare cystic lesion of the liver that is histologically similar to bronchogenic cyst. We report one case of CHFC that was hard to distinguish from solid-cystic neoplasm in imaging features. Magnetic resonance imaging was helpful in differentiating these cysts from other lesions.

  19. Identifying Image Manipulation Software from Image Features


    an overview of the DCT based encoding process [5]. When an image is processed by lossless compression, a file’s size is reduced while still...IDENTIFYING IMAGE MANIPULATION SOFTWARE FROM IMAGE FEATURES THESIS Devlin T. Boyter, CPT, USA AFIT-ENG-MS-15-M-051 DEPARTMENT OF THE AIR FORCE copyright protection in the United States. AFIT-ENG-MS-15-M-051 IDENTIFYING IMAGE MANIPULATION SOFTWARE FROM IMAGE FEATURES THESIS Presented to

  20. Imaging features of thalassemia

    Tunaci, M.; Tunaci, A.; Engin, G.; Oezkorkmaz, B.; Acunas, G.; Acunas, B. [Dept. of Radiology, Istanbul Univ. (Turkey); Dincol, G. [Dept. of Internal Medicine, Istanbul Univ. (Turkey)


    Thalassemia is a kind of chronic, inherited, microcytic anemia characterized by defective hemoglobin synthesis and ineffective erythropoiesis. In all thalassemias clinical features that result from anemia, transfusional, and absorptive iron overload are similar but vary in severity. The radiographic features of {beta}-thalassemia are due in large part to marrow hyperplasia. Markedly expanded marrow space lead to various skeletal manifestations including spine, skull, facial bones, and ribs. Extramedullary hematopoiesis (ExmH), hemosiderosis, and cholelithiasis are among the non-skeletal manifestations of thalassemia. The skeletal X-ray findings show characteristics of chronic overactivity of the marrow. In this article both skeletal and non-skeletal manifestations of thalassemia are discussed with an overview of X-ray findings, including MRI and CT findings. (orig.)

  1. Partial differential equation method based on image feature for Denoising%基于图像特征的偏微分方程去噪方法

    张小华; 王然


    针对不同的自然图像去噪,现有方法的处理结果往往都含有吉布斯效应,目前很难找到非常理想的方法来进行处理。文中提出了一种基于图像特征的偏微分方程图像去噪方法。文中在研究TV模型和PM模型的基础上提出了基于每幅图像具体特征的去噪模型。该模型能够自适应的根据图像每个区域内的细节特征来调节扩散系数的大小,使其能在消除高梯度噪声的同时较好的保留边缘信息。我们证明了该模型的理论性。实验表明改进后的方法在消除噪声的同时也消除了吉布斯现象。%For the special different nature images, we could hardly find particularly desirable approach, and there always exist Gibbs-type artifacts in the results of most methods. A novel Partial Differential Equation (PDE) model is proposed based on image feature for images denoising. The PDE model is adaptive within each region according to the details of the image feature to adjust the size of the diffusion coefficient. So it can be disposed the high gradient noise at the same time better to retain the edge information. We also analyze the performance of the PDE model method. Numerical results show that our algorithm competes favorably with state of the-art TV projection methods to eliminate noise and reduce Gibbs-type artifacts.

  2. Imaging features of aggressive angiomyxoma

    Jeyadevan, N.N.; Sohaib, S.A.A.; Thomas, J.M.; Jeyarajah, A.; Shepherd, J.H.; Fisher, C


    AIM: To describe the imaging features of aggressive angiomyxoma in a rare benign mesenchymal tumour most frequently arising from the perineum in young female patients. MATERIALS AND METHODS: We reviewed the computed tomography (CT) and magnetic resonance (MR) imaging features of patients with aggressive angiomyxoma who were referred to our hospital. The imaging features were correlated with clinical information and pathology in all patients. RESULTS: Four CT and five MR studies were available for five patients (all women, mean age 39, range 24-55). Three patients had recurrent tumour at follow-up. CT and MR imaging demonstrated a well-defined mass-displacing adjacent structures. The tumour was of low attenuation relative to muscle on CT. On MR, the tumour was isointense relative to muscle on T1-weighted image, hyperintense on T2-weighted image and enhanced avidly after gadolinium contrast with a characteristic 'swirled' internal pattern. MR imaging demonstrates the extent of the tumour and its relation to the pelvic floor. Recurrent tumour has a similar appearance to the primary lesion. CONCLUSION: The MR appearances of aggressive angiomyxomas are characteristic, and the diagnosis should be considered in any young woman presenting with a well-defined mass arising from the perineum. Jeyadevan, N. N. etal. (2003). Clinical Radiology58, 157--162.

  3. Featured Image: A Comet's Coma

    Kohler, Susanna


    This series of images (click for the full view!) features the nucleus of comet 67P/Churymov-Gerasimenko. The images were taken with the Wide Angle Camera of RosettasOSIRIS instrument asRosetta orbited comet 67P. Each column represents a different narrow-band filter that allows us to examine the emission of a specific fragment species, and the images progress in time from January 2015 (top) to June 2015 (bottom). In a recent study, Dennis Bodewits (University of Maryland) and collaborators used these images to analyze the comets inner coma, the cloud of gas and dust produced around the nucleus as ices sublime. OSIRISs images allowed the team to explore how the 67Ps inner coma changed over time as the comet approached the Sun marking the first time weve been able to study such an environment at this level of detail. To read more about what Bodewits and collaborators learned, you can check out their paper below!CitationD. Bodewits et al 2016 AJ 152 130. doi:10.3847/0004-6256/152/5/130

  4. Diagnostic Efficacy of All Series of Dynamic Contrast Enhanced Breast MR Images Using Gradient Vector Flow (GVF Segmentation and Novel Border Feature Extraction for Differentiation Between Malignant

    L. Bahreini


    Full Text Available Background/Objective: To discriminate between malignant and benign breast lesions;"nconventionally, the first series of Breast Subtraction Dynamic Contrast-Enhanced Magnetic"nResonance Imaging (BS DCE-MRI images are used for quantitative analysis. In this study, we"ninvestigated whether using all series of these images could provide us with more diagnostic"ninformation."nPatients and Methods: This study included 60 histopathologically proven lesions. The steps of"nthis study were as follows: selecting the regions of interest (ROI, segmentation using Gradient"nVector Flow (GVF snake for the first time, defining new feature sets, using artificial neural network"n(ANN for optimal feature set selection, evaluation using receiver operating characteristic (ROC"nanalysis."nResults: The results showed GVF snake method correctly segmented 95.3% of breast lesion"nborders at the overlap threshold of 0.4. The first classifier which used the optimal feature set"nextracted only from the first series of BS DCE-MRI images achieved an area under the curve"n(AUC of 0.82, specificity of 60% at sensitivity of 81%. The second classifier which used the same"noptimal feature set but was extracted from all five series of these images achieved an AUC of"n0.90, specificity of 79% at sensitivity of 81%."nConclusion: The result of GVF snake segmentation showed that it could make an accurate"nsegmentation in the borders of breast lesions. According to this study, using all five series of BS"nDCE-MRI images could provide us with more diagnostic information about the breast lesion and"ncould improve the performance of breast lesion classifiers in comparison with using the first"nseries alone.

  5. Image feature localization by multiple hypothesis testing of Gabor features.

    Ilonen, Jarmo; Kamarainen, Joni-Kristian; Paalanen, Pekka; Hamouz, Miroslav; Kittler, Josef; Kälviäinen, Heikki


    Several novel and particularly successful object and object category detection and recognition methods based on image features, local descriptions of object appearance, have recently been proposed. The methods are based on a localization of image features and a spatial constellation search over the localized features. The accuracy and reliability of the methods depend on the success of both tasks: image feature localization and spatial constellation model search. In this paper, we present an improved algorithm for image feature localization. The method is based on complex-valued multi resolution Gabor features and their ranking using multiple hypothesis testing. The algorithm provides very accurate local image features over arbitrary scale and rotation. We discuss in detail issues such as selection of filter parameters, confidence measure, and the magnitude versus complex representation, and show on a large test sample how these influence the performance. The versatility and accuracy of the method is demonstrated on two profoundly different challenging problems (faces and license plates).

  6. Multispectral infrared reflectography to differentiate features in paintings.

    Daffara, Claudia; Fontana, Raffaella


    Infrared reflectography is a well-known technique based on wideband imaging in the near-infrared (NIR) range used for painting diagnostics in conservation laboratories.. This work is focused on the application of multiband reflectography for analysis of pictorial layers and differentiated detection of painting features. This technique generates a set of narrowband NIR images of the painting. Starting from a dataset that is registered, metrically correct, and calibrated, the capability of collecting both spectral and spatial information has been exploited by processing the image cube with interplane techniques. Examples on artworks by Caravaggio, Veronese, Bronzino, and Schiavone are presented.

  7. Intramural hemorrhage of the thoracic aorta - imaging features and differential diagnosis; Das intramurale Haematom der thorakalen Aorta: Bildgebende Diagnostik und Differentialdiagnose

    Sommer, T. [Bonn Univ. (Germany) Radiologische Klinik; Abu-Ramadan, D. [Bonn Univ. (Germany). Klinik fuer Herz- und Gefaesschirurgie; Busch, M. [Bonn Univ. (Germany) Radiologische Klinik; Bierhoff, E. [Bonn Univ. (Germany). Pathologische Inst.; Kreft, B. [Bonn Univ. (Germany) Radiologische Klinik; Kuhl, C. [Bonn Univ. (Germany) Radiologische Klinik; Lutterbey, G. [Bonn Univ. (Germany) Radiologische Klinik; Keller, E. [Bonn Univ. (Germany) Radiologische Klinik; Schild, H. [Bonn Univ. (Germany) Radiologische Klinik


    Purpose: Aortic wall thickening due to intramural hemorrhage may be the only sign of aortic dissection. The aim of this study was to evaluate the incidence, imaging features and differential diagnoses of intramural hemorrhage (IMH) of the thoracic aorta. Methods: 98 patients with clinically suspected aortic dissection were investigated via Spiral-CT and MRT. Diagnosis of IMH based on the presence of smooth crescentic or concentric wall thickening over a longer segment of the thoracic aorta without flow visualization and without compression or distortion of the aortic lumen. Results: 69 patients had classic aortic dissections and 7 patients were diagnosed to have IMH of the thoracic aorta. One patient with IMH of the ascending aorta died of aortic rupture and subsequent pericardial tamponade 12 hours after onset of symptoms. In one patient with IMH of the descending aorta on initial examination, there was a progression of overt aortic dissection at follow-up after three weeks. In two patients with IMH of the descending aorta, wall thickening decreased in size at follow-up (10-15 weeks), whereas in one patient it remained unchanged. Conclusion: IMH of the aorta should be considered a precursor of aortic dissection. At follow-up IMH may decrease in size, rupture or progress to overt aortic dissection. (orig.) [Deutsch] Ziel: Eine aortale Wandverdichtung als Ausdruck eines intramuralen Haematoms kann die einzige Manifestation einer Aortendissektion sein. Ziel dieser Arbeit war die Evaluierung der Inzidenz, bildgebenden Aspekte und Differentialdiagnosen dieses in der deutschsprachigen Literatur wenig bekannten Krankheitsbildes. Methode: 98 Patienten mit klinischem Verdacht auf eine Aortendissektion wurden MR- und computertomographisch untersucht. Kriterium fuer das Vorliegen eines intramuralen Haematoms war der Nachweis einer laengerstreckigen aortalen Wandverdickung ohne Flussnachweis sowie ohne Konfigurationsaenderung des aortalen Lumens. Ergebnisse: 69 Patienten

  8. Infrared image mosaic using point feature operators

    Huang, Zhen; Sun, Shaoyuan; Shen, Zhenyi; Hou, Junjie; Zhao, Haitao


    In this paper, we study infrared image mosaic around a single point of rotation, aiming at expanding the narrow view range of infrared images. We propose an infrared image mosaic method using point feature operators including image registration and image synthesis. Traditional mosaic algorithms usually use global image registration methods to extract the feature points in the global image, which cost too much time as well as considerable matching errors. To address this issue, we first roughly calculate the image shift amount using phase correlation and determine the overlap region between images, and then extract image features in overlap region, which shortens the registration time and increases the quality of feature points. We improve the traditional algorithm through increasing constraints of point matching based on prior knowledge of image shift amount based on which the weighted map is computed using fade in-out method. The experimental results verify that the proposed method has better real time performance and robustness.

  9. Medical Image Feature, Extraction, Selection And Classification



    Full Text Available Breast cancer is the most common type of cancer found in women. It is the most frequent form of cancer and one in 22 women in India is likely to suffer from breast cancer. This paper proposes a image classifier to classify the mammogram images. Mammogram image is classified into normal image, benign image and malignant image. Totally 26 features including histogram intensity features and GLCM features are extracted from mammogram image. A hybrid approach of feature selection is proposed in this paper which reduces 75% of the features. Decision tree algorithms are applied to mammography lassification by using these reduced features. Experimental results have been obtained for a data set of 113 images taken from MIAS of different types. This technique of classification has not been attempted before and it reveals the potential of Data mining in medical treatment.

  10. Tongue Image Feature Extraction in TCM

    LI Dong; DU Lian-xiang; LU Fu-ping; DU Jun-ping


    In this paper, digital image processing and computer vision techniques are applied to study tongue images for feature extraction with VC++ and Matlab. Extraction and analysis of the tongue surface features are based on shape, color, edge, and texture. The developed software has various functions and good user interface and is easy to use. Feature data for tongue image pattern recognition is provided, which form a sound basis for the future tongue image recognition.

  11. Perinatal clinical and imaging features of CLOVES syndrome

    Fernandez-Pineda, Israel [Virgen del Rocio Children' s Hospital, Department of Pediatric Surgery, Seville (Spain); Fajardo, Manuel [Virgen del Rocio Children' s Hospital, Department of Pediatric Radiology, Seville (Spain); Chaudry, Gulraiz; Alomari, Ahmad I. [Children' s Hospital Boston and Harvard Medical School, Division of Vascular and Interventional Radiology, Boston, MA (United States)


    We report a neonate with antenatal imaging features suggestive of CLOVES syndrome. Postnatal clinical and imaging findings confirmed the diagnosis, with the constellation of truncal overgrowth, cutaneous capillary malformation, lymphatic and musculoskeletal anomalies. The clinical, radiological and histopathological findings noted in this particular phenotype help differentiate it from other overgrowth syndromes with complex vascular anomalies. (orig.)

  12. Imaging features in Hirayama disease

    Sonwalkar Hemant


    Full Text Available Purpose: To evaluate the MR findings in clinically suspected cases of Hirayama disease. Materials and Methods: The pre and post contrast neutral and flexion position cervical MR images of eight patients of clinically suspected Hirayama disease were evaluated for the following findings: localized lower cervical cord atrophy, asymmetric cord flattening, abnormal cervical curvature, loss of attachment between the posterior dural sac and subjacent lamina, anterior shifting of the posterior wall of the cervical dural canal and enhancing epidural component with flow voids. The distribution of the above features in our patient population was noted and correlated with their clinical presentation and electromyography findings. Observations: Although lower cervical cord atrophy was noted in all eight cases of suspected Hirayama disease, the rest of the findings were variably distributed with asymmetric cord flattening, abnormal cervical curvature, anterior shifting of the posterior wall of the cervical dural canal and enhancing epidural component seen in six out of eight (75% cases. An additional finding of thoracic extension of the enhancing epidural component was also noted in five out of eight cases. Conclusion: Dynamic post contrast MRI evaluation of cervicothoracic spine is an accurate method for the diagnosis of Hirayama disease.

  13. Novel Feature Selection by Differential Evolution Algorithm

    Ali Ghareaghaji


    Full Text Available Iris scan biometrics employs the unique characteristic and features of the human iris in order to verify the identity of in individual. In today's world, where terrorist attacks are on the rise employment of infallible security systems is a must. This makes Iris recognition systems unavoidable in emerging security. Authentication the objective function is minimized using Differential Evolutionary (DE Algorithm where the population vector is encoded using Binary Encoded Decimal to avoid the float number optimization problem. An automatic clustering of the possible values of the Lagrangian multiplier provides a detailed insight of the selected features during the proposed DE based optimization process. The classification accuracy of Support Vector Machine (SVM is used to measure the performance of the selected features. The proposed algorithm outperforms the existing DE based approaches when tested on IRIS, Wine, Wisconsin Breast Cancer, Sonar and Ionosphere datasets. The same algorithm when applied on gait based people identification, using skeleton data points obtained from Microsoft Kinect sensor, exceeds the previously reported accuracies.

  14. Feature-based Image Sequence Compression Coding


    A novel compressing method for video teleconference applications is presented. Semantic-based coding based on human image feature is realized, where human features are adopted as parameters. Model-based coding and the concept of vector coding are combined with the work on image feature extraction to obtain the result.

  15. Average Gait Differential Image Based Human Recognition

    Jinyan Chen


    Full Text Available The difference between adjacent frames of human walking contains useful information for human gait identification. Based on the previous idea a silhouettes difference based human gait recognition method named as average gait differential image (AGDI is proposed in this paper. The AGDI is generated by the accumulation of the silhouettes difference between adjacent frames. The advantage of this method lies in that as a feature image it can preserve both the kinetic and static information of walking. Comparing to gait energy image (GEI, AGDI is more fit to representation the variation of silhouettes during walking. Two-dimensional principal component analysis (2DPCA is used to extract features from the AGDI. Experiments on CASIA dataset show that AGDI has better identification and verification performance than GEI. Comparing to PCA, 2DPCA is a more efficient and less memory storage consumption feature extraction method in gait based recognition.

  16. Magnetic Resonance Imaging Features of Neuromyelitis Optica

    You, Sun Kyung; Song, Chang June; Park, Woon Ju; Lee, In Ho; Son, Eun Hee [Chungnam National University College of Medicine, Chungnam National University Hospital, Daejeon (Korea, Republic of)


    To report the magnetic resonance (MR) imaging features of the spinal cord and brain in patients of neuromyelitis optica (NMO). Between January 2001 and March 2010, the MR images (spinal cord, brain, and orbit) and the clinical and serologic findings of 11 NMO patients were retrospectively reviewed. The contrast-enhancement of the spinal cord was performed (20/23). The presence and pattern of the contrast-enhancement in the spinal cord were classified into 5 types. Acute myelitis was monophasic in 8 patients (8/11, 72.7%); and optic neuritis preceded acute myelitis in most patients. Longitudinally extensive cord lesion (average, 7.3 vertebral segments) was involved. The most common type was the diffuse and subtle enhancement of the spinal cord with a multifocal nodular, linear or segmental intense enhancement (45%). Most of the brain lesions (5/11, 10 lesions) were located in the brain stem, thalamus and callososeptal interphase. Anti-Ro autoantibody was positive in 2 patients, and they showed a high relapse rate of acute myelitis. Anti-NMO IgG was positive in 4 patients (4/7, 66.7%). The imaging findings of acute myelitis in NMO may helpful in making an early diagnosis of NMO which can result in a severe damage to the spinal cord, and to make a differential diagnosis of multiple sclerosis and inflammatory diseases of the spinal cord such as toxocariasis.

  17. Hepatic CT image query using Gabor features

    Chenguang Zhao(赵晨光); Hongyan Cheng(程红岩); Tiange Zhuang(庄天戈)


    A retrieval scheme for liver computerize tomography (CT) images based on Gabor texture is presented.For each hepatic CT image, we manually delineate abnormal regions within liver area. Then, a continuous Gabor transform is utilized to analyze the texture of the pathology bearing region and extract the corresponding feature vectors. For a given sample image, we compare its feature vector with those of other images. Similar images with the highest rank are retrieved. In experiments, 45 liver CT images are collected, and the effectiveness of Gabor texture for content based retrieval is verified.

  18. Differential morphology and image processing.

    Maragos, P


    Image processing via mathematical morphology has traditionally used geometry to intuitively understand morphological signal operators and set or lattice algebra to analyze them in the space domain. We provide a unified view and analytic tools for morphological image processing that is based on ideas from differential calculus and dynamical systems. This includes ideas on using partial differential or difference equations (PDEs) to model distance propagation or nonlinear multiscale processes in images. We briefly review some nonlinear difference equations that implement discrete distance transforms and relate them to numerical solutions of the eikonal equation of optics. We also review some nonlinear PDEs that model the evolution of multiscale morphological operators and use morphological derivatives. Among the new ideas presented, we develop some general 2-D max/min-sum difference equations that model the space dynamics of 2-D morphological systems (including the distance computations) and some nonlinear signal transforms, called slope transforms, that can analyze these systems in a transform domain in ways conceptually similar to the application of Fourier transforms to linear systems. Thus, distance transforms are shown to be bandpass slope filters. We view the analysis of the multiscale morphological PDEs and of the eikonal PDE solved via weighted distance transforms as a unified area in nonlinear image processing, which we call differential morphology, and briefly discuss its potential applications to image processing and computer vision.

  19. Wilson’s disease: Atypical imaging features

    Venugopalan Y Vishnu


    Full Text Available Wilson’s disease is a genetic movement disorder with characteristic clinical and imaging features. We report a 17- year-old boy who presented with sialorrhea, hypophonic speech, paraparesis with repeated falls and recurrent seizures along with cognitive decline. He had bilateral Kayser Flescher rings. Other than the typical features of Wilson’s disease in cranial MRI, there were extensive white matter signal abnormalities (T2 and FLAIR hyperintensities and gyriform contrast enhancement which are rare imaging features in Wilson's disease. A high index of suspicion is required to diagnose Wilson’s disease when atypical imaging features are present.

  20. Image segmentation using association rule features.

    Rushing, John A; Ranganath, Heggere; Hinke, Thomas H; Graves, Sara J


    A new type of texture feature based on association rules is described. Association rules have been used in applications such as market basket analysis to capture relationships present among items in large data sets. It is shown that association rules can be adapted to capture frequently occurring local structures in images. The frequency of occurrence of these structures can be used to characterize texture. Methods for segmentation of textured images based on association rule features are described. Simulation results using images consisting of man made and natural textures show that association rule features perform well compared to other widely used texture features. Association rule features are used to detect cumulus cloud fields in GOES satellite images and are found to achieve higher accuracy than other statistical texture features for this problem.

  1. Image fusion using sparse overcomplete feature dictionaries

    Brumby, Steven P.; Bettencourt, Luis; Kenyon, Garrett T.; Chartrand, Rick; Wohlberg, Brendt


    Approaches for deciding what individuals in a population of visual system "neurons" are looking for using sparse overcomplete feature dictionaries are provided. A sparse overcomplete feature dictionary may be learned for an image dataset and a local sparse representation of the image dataset may be built using the learned feature dictionary. A local maximum pooling operation may be applied on the local sparse representation to produce a translation-tolerant representation of the image dataset. An object may then be classified and/or clustered within the translation-tolerant representation of the image dataset using a supervised classification algorithm and/or an unsupervised clustering algorithm.

  2. Imaging features of iliopsoas bursitis

    Wunderbaldinger, P. [Department of Radiology, University of Vienna (Austria); Center of Molecular Imaging Research, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA (United States); Bremer, C. [Department of Radiology, University of Muenster (Germany); Schellenberger, E. [Center of Molecular Imaging Research, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA (United States); Department of Radiology, Martin-Luther University of Halle-Wittenberg, Halle (Germany); Cejna, M.; Turetschek, K.; Kainberger, F. [Department of Radiology, University of Vienna (Austria)


    The aim of this study was firstly to describe the spectrum of imaging findings seen in iliopsoas bursitis, and secondly to compare cross-sectional imaging techniques in the demonstration of the extent, size and appearance of the iliopsoas bursitis as referenced by surgery. Imaging studies of 18 patients (13 women, 5 men; mean age 53 years) with surgically proven iliopsoas bursitis were reviewed. All patients received conventional radiographs of the pelvis and hip, US and MR imaging of the hip. The CT was performed in 5 of the 18 patients. Ultrasound, CT and MR all demonstrated enlarged iliopsoas bursae. The bursal wall was thin and well defined in 83% and thickened in 17% of all cases. The two cases with septations on US were not seen by CT and MRI. A communication between the bursa and the hip joint was seen, and surgically verified, in all 18 patients by MR imaging, whereas US and CT failed to demonstrate it in 44 and 40% of the cases, respectively. Hip joint effusion was seen and verified by surgery in 16 patients by MRI, whereas CT (4 of 5) and US (n=12) underestimated the number. The overall size of the bursa corresponded best between MRI and surgery, whereas CT and US tended to underestimate the size. Contrast enhancement of the bursal wall was seen in all cases. The imaging characteristics of iliopsoas bursitis are a well-defined, thin-walled cystic mass with a communication to the hip joint and peripheral contrast enhancement. The most accurate way to assess iliopsoas bursitis is with MR imaging; thus, it should be used for accurate therapy planning and follow-up studies. In order to initially prove an iliopsoas bursitis, US is the most cost-effective, easy-to-perform and fast alternative. (orig.)


    Vitaly V. Bezzubik


    Full Text Available We have proposed and tested a novel method for digital image sharpening. The method is based on multi-scale image analysis, calculation of differential responses of image brightness in different spatial scales, and the subsequent calculation of a restoration function, which sharpens the image by simple subtraction of its brightness values from those of the original image. The method features spatial transposition of the restoration function elements, its normalization, and taking into account the sign of the brightness differential response gradient close to the object edges. The calculation algorithm for the proposed method makes use of integer arithmetic that significantly reduces the computation time. The paper shows that for the images containing small amount of the blur due to the residual aberrations of an imaging system, only the first two scales are needed for the calculation of the restoration function. Similar to the blind deconvolution, the method requires no a priori information about the nature and magnitude of the blur kernel, but it is computationally inexpensive and is much easier in practical implementation. The most promising applications of the method are machine vision and surveillance systems based on real-time intelligent pattern recognition and decision making.

  4. Content Based Image Retrieval by Multi Features using Image Blocks

    Arpita Mathur


    Full Text Available Content based image retrieval (CBIR is an effective method of retrieving images from large image resources. CBIR is a technique in which images are indexed by extracting their low level features like, color, texture, shape, and spatial location, etc. Effective and efficient feature extraction mechanisms are required to improve existing CBIR performance. This paper presents a novel approach of CBIR system in which higher retrieval efficiency is achieved by combining the information of image features color, shape and texture. The color feature is extracted using color histogram for image blocks, for shape feature Canny edge detection algorithm is used and the HSB extraction in blocks is used for texture feature extraction. The feature set of the query image are compared with the feature set of each image in the database. The experiments show that the fusion of multiple features retrieval gives better retrieval results than another approach used by Rao et al. This paper presents comparative study of performance of the two different approaches of CBIR system in which the image features color, shape and texture are used.

  5. Finding curvilinear features in speckled images

    Samadani, Ramin; Vesecky, John F.


    A method for finding curves in digital images with speckle noise is described. The solution method differs from standard linear convolutions followed by thresholds in that it explicitly allows curvature in the features. Maximum a posteriori (MAP) estimation is used, together with statistical models for the speckle noise and for the curve-generation process, to find the most probable estimate of the feature, given the image data. The estimation process is first described in general terms. Then, incorporation of the specific neighborhood system and a multiplicative noise model for speckle allows derivation of the solution, using dynamic programming, of the estimation problem. The detection of curvilinear features is considered separately. The detection results allow the determination of the minimal size of detectable feature. Finally, the estimation of linear features, followed by a detection step, is shown for computer-simulated images and for a SAR image of sea ice.

  6. Multi Feature Content Based Image Retrieval

    Rajshree S. Dubey,


    Full Text Available There are numbers of methods prevailing for Image Mining Techniques. This Paper includes the features of four techniques I,e Color Histogram, Color moment, Texture, and Edge Histogram Descriptor. The nature of the Image is basically based on the Human Perception of the Image. The Machine interpretation of the Image is based on the Contours and surfaces of the Images. The study of the Image Mining is a very challenging task because it involves the Pattern Recognition which is a very important tool for the Machine Vision system. A combination of four feature extraction methods namely color istogram, Color Moment, texture, and Edge Histogram Descriptor. There is a provision to add new features in future for better retrievalefficiency. In this paper the combination of the four techniques are used and the Euclidian distances are calculated of the every features are added and the averages are made .The user interface is provided by the Mat lab. The image properties analyzed in this work are by using computer vision and image processing algorithms. For colorthe histogram of images are computed, for texture co occurrence matrix based entropy, energy, etc, are calculated and for edge density it is Edge Histogram Descriptor (EHD that is found. For retrieval of images, the averages of the four techniques are made and the resultant Image is retrieved.

  7. Featured Image: Identifying Weird Galaxies

    Kohler, Susanna


    Hoags Object, an example of a ring galaxy. [NASA/Hubble Heritage Team/Ray A. Lucas (STScI/AURA)]The above image (click for the full view) shows PanSTARRSobservationsof some of the 185 galaxies identified in a recent study as ring galaxies bizarre and rare irregular galaxies that exhibit stars and gas in a ring around a central nucleus. Ring galaxies could be formed in a number of ways; one theory is that some might form in a galaxy collision when a smaller galaxy punches through the center of a larger one, triggering star formation around the center. In a recent study, Ian Timmis and Lior Shamir of Lawrence Technological University in Michigan explore ways that we may be able to identify ring galaxies in the overwhelming number of images expected from large upcoming surveys. They develop a computer analysis method that automatically finds ring galaxy candidates based on their visual appearance, and they test their approach on the 3 million galaxy images from the first PanSTARRS data release. To see more of the remarkable galaxies the authors found and to learn more about their identification method, check out the paper below.CitationIan Timmis and Lior Shamir 2017 ApJS 231 2. doi:10.3847/1538-4365/aa78a3

  8. Quantitative imaging features: extension of the oncology medical image database

    Patel, M. N.; Looney, P. T.; Young, K. C.; Halling-Brown, M. D.


    Radiological imaging is fundamental within the healthcare industry and has become routinely adopted for diagnosis, disease monitoring and treatment planning. With the advent of digital imaging modalities and the rapid growth in both diagnostic and therapeutic imaging, the ability to be able to harness this large influx of data is of paramount importance. The Oncology Medical Image Database (OMI-DB) was created to provide a centralized, fully annotated dataset for research. The database contains both processed and unprocessed images, associated data, and annotations and where applicable expert determined ground truths describing features of interest. Medical imaging provides the ability to detect and localize many changes that are important to determine whether a disease is present or a therapy is effective by depicting alterations in anatomic, physiologic, biochemical or molecular processes. Quantitative imaging features are sensitive, specific, accurate and reproducible imaging measures of these changes. Here, we describe an extension to the OMI-DB whereby a range of imaging features and descriptors are pre-calculated using a high throughput approach. The ability to calculate multiple imaging features and data from the acquired images would be valuable and facilitate further research applications investigating detection, prognosis, and classification. The resultant data store contains more than 10 million quantitative features as well as features derived from CAD predictions. Theses data can be used to build predictive models to aid image classification, treatment response assessment as well as to identify prognostic imaging biomarkers.

  9. Extraction of Facial Features from Color Images

    J. Pavlovicova


    Full Text Available In this paper, a method for localization and extraction of faces and characteristic facial features such as eyes, mouth and face boundaries from color image data is proposed. This approach exploits color properties of human skin to localize image regions – face candidates. The facial features extraction is performed only on preselected face-candidate regions. Likewise, for eyes and mouth localization color information and local contrast around eyes are used. The ellipse of face boundary is determined using gradient image and Hough transform. Algorithm was tested on image database Feret.

  10. An Image Retrieval Method Using DCT Features

    樊昀; 王润生


    In this paper a new image representation for compressed domain image re-trieval and an image retrieval system are presented. To represent images compactly and hi-erarchically, multiple features such as color and texture features directly extracted from DCTcoefficients are structurally organized using vector quantization. To train the codebook, a newMinimum Description Length vector quantization algorithm is used and it automatically decidesthe number of code words. To compare two images using the proposed representation, a newefficient similarity measure is designed. The new method is applied to an image database with1,005 pictures. The results demonstrate that the method is better than two typical histogrammethods and two DCT-based image retrieval methods.

  11. Featured Image: Modeling Supernova Remnants

    Kohler, Susanna


    This image shows a computer simulation of the hydrodynamics within a supernova remnant. The mixing between the outer layers (where color represents the log of density) is caused by turbulence from the Rayleigh-Taylor instability, an effect that arises when the expanding core gas of the supernova is accelerated into denser shell gas. The past standard for supernova-evolution simulations was to perform them in one dimension and then, in post-processing, manually smooth out regions that undergo Rayleigh-Taylor turbulence (an intrinsically multidimensional effect). But in a recent study, Paul Duffell (University of California, Berkeley) has explored how a 1D model could be used to reproduce the multidimensional dynamics that occur in turbulence from this instability. For more information, check out the paper below!CitationPaul C. Duffell 2016 ApJ 821 76. doi:10.3847/0004-637X/821/2/76

  12. Imaging Features and Differential Diagnosis of Posterior Reversible Encephalopathy Syndrome Attributed to Hypertension During Pregnancy%妊高症致后循环脑病的影像特征与鉴别诊断

    李福彰; 孙多成; 靳瑞娟; 肖飞鹰


    目的 探讨大脑后部可逆性脑病综合征( PRES)的影像特征及鉴别诊断.方法 回顾分析35例子痫患者影像资料.其中20例有治疗后复查的影像资料,15例有治疗过程中的MRI检查资料.结果 26例CT表现为顶和(或)枕叶片状低密度,且具有可逆性;15例MR表现为T1WI稍低信号,T2WI稍高信号,T2Flair呈高信号,DWI呈等信号,另7例合并脑出血.结论 大脑后部可逆性脑病较具有特征性,但需要与其他脑内片状水肿影像表现的疾病进行鉴别.%Obiectivelo investigate imaging features and differential diagnosis of Posterior reversible encephalopathy syndrome attributed to Hypertension during pregnancy. Methods The imaging findings of 35 patients with eclampsia were analyzed retrospectively (20 cases had images after treatment; 15patients had MRI images during treatment). Results CT and MRI images of 26 cases showed patchy low density area in the parietal and occipital lobe which is reversible. 15 cases showed slightly low signal intensity on TjWI, slightly high signal intensity on T2WI, high signal intensity on T2Flair and equal signal intensity on DWI. Images of other 7 cases showed combined brain hemorrhage. Conclusion Images of PRES have clear characteristics, but it must be distinguished from other diseases which showed intracephalic pachy edema.

  13. MR imaging features of craniodiaphyseal dysplasia

    Marden, Franklin A. [Mallinckrodt Institute of Radiology, Washington University Medical Center, 510 South Kingshighway Blvd., MO 63110, St. Louis (United States); Department of Radiology, St. Louis Children' s Hospital, Children' s Place, MO 63110, St. Louis (United States); Wippold, Franz J. [Mallinckrodt Institute of Radiology, Washington University Medical Center, 510 South Kingshighway Blvd., MO 63110, St. Louis (United States); Department of Radiology, St. Louis Children' s Hospital, Children' s Place, MO 63110, St. Louis (United States); Department of Radiology/Nuclear Medicine, F. Edward Hebert School of Medicine, Uniformed Services University of the Health Sciences, MD 20814, Bethesda (United States)


    We report the magnetic resonance (MR) imaging findings in a 4-year-old girl with characteristic radiographic and computed tomography (CT) features of craniodiaphyseal dysplasia. MR imaging exquisitely depicted cranial nerve compression, small foramen magnum, hydrocephalus, and other intracranial complications of this syndrome. A syrinx of the cervical spinal cord was demonstrated. We suggest that MR imaging become a routine component of the evaluation of these patients. (orig.)

  14. Solving jigsaw puzzles using image features

    Nielsen, Ture R.; Drewsen, Peter; Hansen, Klaus


    algorithm which exploits the divide and conquer paradigm to reduce the combinatorially complex problem by classifying the puzzle pieces and comparing pieces drawn from the same group. The paper includes a brief preliminary investigation of some image features used in the classification.......In this article, we describe a method for automatic solving of the jigsaw puzzle problem based on using image features instead of the shape of the pieces. The image features are used for obtaining an accurate measure for edge similarity to be used in a new edge matching algorithm. The algorithm...... is used in a general puzzle solving method which is based on a greedy algorithm previously proved successful. We have been able to solve computer generated puzzles of 320 pieces as well as a real puzzle of 54 pieces by exclusively using image information. Additionally, we investigate a new scalable...

  15. Exact feature probabilities in images with occlusion

    Pitkow, Xaq


    To understand the computations of our visual system, it is important to understand also the natural environment it evolved to interpret. Unfortunately, existing models of the visual environment are either unrealistic or too complex for mathematical description. Here we describe a naturalistic image model and present a mathematical solution for the statistical relationships between the image features and model variables. The world described by this model is composed of independent, opaque, textured objects which occlude each other. This simple structure allows us to calculate the joint probability distribution of image values sampled at multiple arbitrarily located points, without approximation. This result can be converted into probabilistic relationships between observable image features as well as between the unobservable properties that caused these features, including object boundaries and relative depth. Using these results we explain the causes of a wide range of natural scene properties, including high...

  16. Automatic extraction of planetary image features

    LeMoigne-Stewart, Jacqueline J. (Inventor); Troglio, Giulia (Inventor); Benediktsson, Jon A. (Inventor); Serpico, Sebastiano B. (Inventor); Moser, Gabriele (Inventor)


    A method for the extraction of Lunar data and/or planetary features is provided. The feature extraction method can include one or more image processing techniques, including, but not limited to, a watershed segmentation and/or the generalized Hough Transform. According to some embodiments, the feature extraction method can include extracting features, such as, small rocks. According to some embodiments, small rocks can be extracted by applying a watershed segmentation algorithm to the Canny gradient. According to some embodiments, applying a watershed segmentation algorithm to the Canny gradient can allow regions that appear as close contours in the gradient to be segmented.

  17. Hemorrhage detection in MRI brain images using images features

    Moraru, Luminita; Moldovanu, Simona; Bibicu, Dorin; Stratulat (Visan), Mirela


    The abnormalities appear frequently on Magnetic Resonance Images (MRI) of brain in elderly patients presenting either stroke or cognitive impairment. Detection of brain hemorrhage lesions in MRI is an important but very time-consuming task. This research aims to develop a method to extract brain tissue features from T2-weighted MR images of the brain using a selection of the most valuable texture features in order to discriminate between normal and affected areas of the brain. Due to textural similarity between normal and affected areas in brain MR images these operation are very challenging. A trauma may cause microstructural changes, which are not necessarily perceptible by visual inspection, but they could be detected by using a texture analysis. The proposed analysis is developed in five steps: i) in the pre-processing step: the de-noising operation is performed using the Daubechies wavelets; ii) the original images were transformed in image features using the first order descriptors; iii) the regions of interest (ROIs) were cropped from images feature following up the axial symmetry properties with respect to the mid - sagittal plan; iv) the variation in the measurement of features was quantified using the two descriptors of the co-occurrence matrix, namely energy and homogeneity; v) finally, the meaningful of the image features is analyzed by using the t-test method. P-value has been applied to the pair of features in order to measure they efficacy.

  18. Wood recognition using image texture features.

    Hang-jun Wang

    Full Text Available Inspired by theories of higher local order autocorrelation (HLAC, this paper presents a simple, novel, yet very powerful approach for wood recognition. The method is suitable for wood database applications, which are of great importance in wood related industries and administrations. At the feature extraction stage, a set of features is extracted from Mask Matching Image (MMI. The MMI features preserve the mask matching information gathered from the HLAC methods. The texture information in the image can then be accurately extracted from the statistical and geometrical features. In particular, richer information and enhanced discriminative power is achieved through the length histogram, a new histogram that embodies the width and height histograms. The performance of the proposed approach is compared to the state-of-the-art HLAC approaches using the wood stereogram dataset ZAFU WS 24. By conducting extensive experiments on ZAFU WS 24, we show that our approach significantly improves the classification accuracy.

  19. Onboard Image Registration from Invariant Features

    Wang, Yi; Ng, Justin; Garay, Michael J.; Burl, Michael C


    This paper describes a feature-based image registration technique that is potentially well-suited for onboard deployment. The overall goal is to provide a fast, robust method for dynamically combining observations from multiple platforms into sensors webs that respond quickly to short-lived events and provide rich observations of objects that evolve in space and time. The approach, which has enjoyed considerable success in mainstream computer vision applications, uses invariant SIFT descriptors extracted at image interest points together with the RANSAC algorithm to robustly estimate transformation parameters that relate one image to another. Experimental results for two satellite image registration tasks are presented: (1) automatic registration of images from the MODIS instrument on Terra to the MODIS instrument on Aqua and (2) automatic stabilization of a multi-day sequence of GOES-West images collected during the October 2007 Southern California wildfires.

  20. Imaging features of alveolar soft part sarcoma

    Teng Jin; Ping Zhang Co-first author; Xiaoming Li


    Objective The aim of this study was to analyze the imaging features of alveolar soft part sarcoma (ASPS). Methods The imaging features of 11 cases with ASPS were retrospectively analyzed. Results ASPS mainly exhibited an isointense or slightly high signal intensity on T1-weighted imaging (T1WI), and a mixed high signal on T2-weighted imaging (T2WI). ASPS was partial, with rich tortuous flow voids, or “line-like” low signal septa. The essence of the mass was heterogeneous enhancement. The 1H-MRS showed a slight choline peak at 3.2 ppm. Conclusion The wel-circumscribed mass and blood voids, combined with “line-like” low signals play a significant role in diagnosis. The choline peak and the other signs may be auxiliary diagnoses.

  1. The role of molecular imaging in the differential diagnosis of parkinsonism

    Booij, J.; Teune, L. K.; Verberne, H. J.


    This review focuses on the possibilities to use scintigraphic techniques to help differentiate neurodegenerative brain diseases associated with parkinsonian features. In particular, the findings of dopaminergic imaging, FDG PET imaging, and cardiac sympathetic imaging are described. Considerable ove

  2. Straight line feature based image distortion correction

    Zhang Haofeng; Zhao Chunxia; Lu Jianfeng; Tang Zhenmin; Yang Jingyu


    An image distortion correction method is proposed, which uses the straight line features. Many parallel lines of different direction from different images were extracted, and then were used to optimize the distortion parameters by nonlinear least square. The thought of step by step was added when the optimization method working. 3D world coordi-nation is not need to know, and the method is easy to implement. The experiment result shows its high accuracy.

  3. [Differential personality features in adult ADHD subtypes].

    Martínez Ortega, Yolanda; Bosch Munsó, Rosa; Gomà-i-Freixanet, Montserrat; Valero Ventura, Sergi; Ramos-Quiroga, Josep Antoni; Nogueira, Mariana; Casas Brugué, Miguel


    Attention Deficit/Hyperactivity Disorder (ADHD) and personality traits are relatively stable from childhood and across life span. The purpose of this study was to identify differential and discriminative personality traits between clinical subtypes of ADHD in adults. The Zuckerman-Kuhlman Personality Questionnaire (ZKPQ) and the Millon Multiaxial Clinical Inventory-II (MCMI-II) were administered to a sample of 146 adults with ADHD. Activity and Aggression-Hostility dimensions from the ZKPQ allowed us to properly classify 75.8% of the inattentive and combined subtypes. Data indicates that ADHD is not a homogeneous entity, but rather, there are significant differences in personality characteristics among clinical subtypes. The results have theoretical implications about the connection between ADHD and personality, and clinical implications regarding diagnosis and treatment designs better tailored to the characteristics of each subtype.

  4. Image Mining Using Texture and Shape Feature

    Prof.Rupali Sawant


    Full Text Available Discovering knowledge from data stored in typical alphanumeric databases, such as relational databases, has been the focal point of most of the work in database mining. However, with advances in secondary and tertiary storage capacity, coupled with a relatively low storage cost, more and more non standard data (in the form of images is being accumulated. This vast collection of image data can also be mined to discover new and valuable knowledge. During theprocess of image mining, the concepts in different hierarchiesand their relationships are extracted from different hierarchies and granularities, and association rule mining and concept clustering are consequently implemented. The generalization and specialization of concepts are realized in different hierarchies, lower layer concepts can be upgraded to upper layer concepts, and upper layer concepts guide the extraction of lower layer concepts. It is a process from image data to image information, from image information to imageknowledge, from lower layer concepts to upper layer concept lattice and cloud model theory is proposed. The methods of image mining from image texture and shape features are introduced here, which include the following basic steps: firstly pre-process images secondly use cloud model to extract concepts, lastly use concept lattice to extracta series of image knowledge.

  5. Imaging features of benign adrenal cysts

    Sanal, Hatice Tuba [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey)]. E-mail:; Kocaoglu, Murat [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey); Yildirim, Duzgun [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey); Bulakbasi, Nail [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey); Guvenc, Inanc [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey); Tayfun, Cem [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey); Ucoz, Taner [Department of Radiology, Gulhane Military Medical Academy, Ankara (Turkey)


    Benign adrenal gland cysts (BACs) are rare lesions with a variable histological spectrum and may mimic not only each other but also malignant ones. We aimed to review imaging features of BACs which can be helpful in distinguishing each entity and determining the subsequent appropriate management.

  6. Disorders of cortical formation: MR imaging features.

    Abdel Razek, A A K; Kandell, A Y; Elsorogy, L G; Elmongy, A; Basett, A A


    The purpose of this article was to review the embryologic stages of the cerebral cortex, illustrate the classification of disorders of cortical formation, and finally describe the main MR imaging features of these disorders. Disorders of cortical formation are classified according to the embryologic stage of the cerebral cortex at which the abnormality occurred. MR imaging shows diminished cortical thickness and sulcation in microcephaly, enlarged dysplastic cortex in hemimegalencephaly, and ipsilateral focal cortical thickening with radial hyperintense bands in focal cortical dysplasia. MR imaging detects smooth brain in classic lissencephaly, the nodular cortex with cobblestone cortex with congenital muscular dystrophy, and the ectopic position of the gray matter with heterotopias. MR imaging can detect polymicrogyria and related syndromes as well as the types of schizencephaly. We concluded that MR imaging is essential to demonstrate the morphology, distribution, and extent of different disorders of cortical formation as well as the associated anomalies and related syndromes.

  7. Focal hepatic lesions with peripheral eosinophilia: imaging features of various disease

    Seo, Joon Beom; Han, Joon Koo; Kim, Tae Kyoung; Choi, Byung Ihn; Han, Man Chung [Seoul National Univ. College of Medicine, Seoul (Korea, Republic of); Song, Chi Sung [Seoul City Boramae Hospital, Seoul (Korea, Republic of)


    Due to the recent advent of various imaging modalities such as ultrasonography, computed tomography and magnetic resonance imaging, as well as knowledge of the characteristic imaging features of hepatic lesions, radiologic examination plays a major role in the differential diagnosis of focal hepatic lesions. However, various 'nonspecific' or 'unusual' imaging features of focal hepatic lesions are occasionally encountered, and this makes correct diagnosis difficult. In such a situation, the presence of peripheral eosinophilia helps narrow the differential diagnoses. The aim of this pictorial essay is to describe the imaging features of various disease entities which cause focal hepatic lesions and peripheral eosinophilia.

  8. Clinical and imaging features of fludarabine neurotoxicity.

    Lee, Michael S; McKinney, Alexander M; Brace, Jeffrey R; Santacruz, Karen


    Neurotoxicity from intravenous fludarabine is a rare but recognized clinical entity. Its brain imaging features have not been extensively described. Three patients received 38.5 mg or 40 mg/m per day fludarabine in a 5-day intravenous infusion before bone marrow transplantation in treatment of hematopoietic malignancies. Several weeks later, each patient developed progressive neurologic decline, including retrogeniculate blindness, leading to coma and death. Brain MRI showed progressively enlarging but mild T2/FLAIR hyperintensities in the periventricular white matter. The lesions demonstrated restricted diffusion but did not enhance. Because the neurotoxicity of fludarabine appears long after exposure, neurologic decline in this setting is likely to be attributed to opportunistic disease. However, the imaging features are distinctive in their latency and in being mild relative to the profound clinical features. The safe dose of fludarabine in this context remains controversial.

  9. Ophthalmic imaging features of posterior scleritis

    Zhi Li


    Full Text Available AIM: To analyze, summarize and describe ophthalmic imaging features of posterior scleritis. METHODS: Clinical data of 16 patients(21 eyeswith posterior scleritis diagnosed in our hospital from October 2008 to June 2013 were retrospectively analyzed. The results of type-B ultrasonic, fundus chromophotograph, fundus fluorescein angiography, CT were recorded for comprehensive evaluation and analysis of ophthalmic imaging features of posterior scleritis. RESULTS: All patients underwent type-B ultrasonic examination and manifested as diffuse and nodular types. The diffuse type showed diffusely thickened sclera and a dark hypoechoic area that connected with the optic nerve to form a typical “T”-shaped sign. The nodular type showed scleral echogenic nodules and relatively regular internal structure. FFA showed that relatively weak mottled fluorescences were visible in the arterial early phase and strong multiple needle-like fluorescences were visible in the arteriovenous phase, which were then progressively larger and fused; fluorescein was leaked to the subretinal tissue in the late phase; varying degrees of strong fluorescences with less clear or unclear boundaries were visible in the optic disk. CT results showed thickened eyeball wall. CONCLUSION: Posterior scleritis is common in young female patients, whose ophthalmic imaging features are varied and more specific in type-B ultrasonic. Selection of rational ophthalmic imaging examination method, combined with clinical manifestations, can accurately diagnose posterior scleritis and avoid the incidence of missed and delayed diagnosis.

  10. Multispectral image fusion based on fractal features

    Tian, Jie; Chen, Jie; Zhang, Chunhua


    Imagery sensors have been one indispensable part of the detection and recognition systems. They are widely used to the field of surveillance, navigation, control and guide, et. However, different imagery sensors depend on diverse imaging mechanisms, and work within diverse range of spectrum. They also perform diverse functions and have diverse circumstance requires. So it is unpractical to accomplish the task of detection or recognition with a single imagery sensor under the conditions of different circumstances, different backgrounds and different targets. Fortunately, the multi-sensor image fusion technique emerged as important route to solve this problem. So image fusion has been one of the main technical routines used to detect and recognize objects from images. While, loss of information is unavoidable during fusion process, so it is always a very important content of image fusion how to preserve the useful information to the utmost. That is to say, it should be taken into account before designing the fusion schemes how to avoid the loss of useful information or how to preserve the features helpful to the detection. In consideration of these issues and the fact that most detection problems are actually to distinguish man-made objects from natural background, a fractal-based multi-spectral fusion algorithm has been proposed in this paper aiming at the recognition of battlefield targets in the complicated backgrounds. According to this algorithm, source images are firstly orthogonally decomposed according to wavelet transform theories, and then fractal-based detection is held to each decomposed image. At this step, natural background and man-made targets are distinguished by use of fractal models that can well imitate natural objects. Special fusion operators are employed during the fusion of area that contains man-made targets so that useful information could be preserved and features of targets could be extruded. The final fused image is reconstructed from the

  11. Barrett's esophagus: clinical features, obesity, and imaging.

    Quigley, Eamonn M M


    The following includes commentaries on clinical features and imaging of Barrett\\'s esophagus (BE); the clinical factors that influence the development of BE; the influence of body fat distribution and central obesity; the role of adipocytokines and proinflammatory markers in carcinogenesis; the role of body mass index (BMI) in healing of Barrett\\'s epithelium; the role of surgery in prevention of carcinogenesis in BE; the importance of double-contrast esophagography and cross-sectional images of the esophagus; and the value of positron emission tomography\\/computed tomography.

  12. Special feature on imaging systems and techniques

    Yang, Wuqiang; Giakos, George


    The IEEE International Conference on Imaging Systems and Techniques (IST'2012) was held in Manchester, UK, on 16-17 July 2012. The participants came from 26 countries or regions: Austria, Brazil, Canada, China, Denmark, France, Germany, Greece, India, Iran, Iraq, Italy, Japan, Korea, Latvia, Malaysia, Norway, Poland, Portugal, Sweden, Switzerland, Taiwan, Tunisia, UAE, UK and USA. The technical program of the conference consisted of a series of scientific and technical sessions, exploring physical principles, engineering and applications of new imaging systems and techniques, as reflected by the diversity of the submitted papers. Following a rigorous review process, a total of 123 papers were accepted, and they were organized into 30 oral presentation sessions and a poster session. In addition, six invited keynotes were arranged. The conference not only provided the participants with a unique opportunity to exchange ideas and disseminate research outcomes but also paved a way to establish global collaboration. Following the IST'2012, a total of 55 papers, which were technically extended substantially from their versions in the conference proceeding, were submitted as regular papers to this special feature of Measurement Science and Technology . Following a rigorous reviewing process, 25 papers have been finally accepted for publication in this special feature and they are organized into three categories: (1) industrial tomography, (2) imaging systems and techniques and (3) image processing. These papers not only present the latest developments in the field of imaging systems and techniques but also offer potential solutions to existing problems. We hope that this special feature provides a good reference for researchers who are active in the field and will serve as a catalyst to trigger further research. It has been our great pleasure to be the guest editors of this special feature. We would like to thank the authors for their contributions, without which it would

  13. Osteosarcoma of pelvic bones: imaging features.

    Park, Se Kyoung; Lee, In Sook; Cho, Kil Ho; Lee, Young Hwan; Yi, Jae Hyuck; Choi, Kyung Un

    The metaphyseal locations of tubular bones with osteoid mineralization in young patients are important diagnostic radiologic features of osteosarcoma. The pelvic bones are an unusual location of osteosarcoma. Although osteosarcoma occurring in pelvic bones is not common, the osteoid matrix may be a critical finding for differentiating osteosarcoma from other common pelvic bone tumors. Therefore, the possibility of osteosarcoma in pelvic bones may be considered in the presence of osteoid matrix even in the old age group. Copyright © 2016. Published by Elsevier Inc.

  14. Imaging features of juxtacortical chondroma in children

    Miller, Stephen F. [St. Jude Children' s Research Hospital, Department of Radiological Sciences, Memphis, TN (United States)


    Juxtacortical chondroma is a rare benign bone lesion in children. Children usually present with a mildly painful mass, which prompts diagnostic imaging studies. The rarity of this condition often presents a diagnostic challenge. Correct diagnosis is crucial in guiding surgical management. To describe the characteristic imaging findings of juxtacortical chondroma in children. We identified all children who were diagnosed with juxtacortical chondroma between 1998 and 2012. A single experienced pediatric radiologist reviewed all diagnostic imaging studies, including plain radiographs, CT, MR and bone scans. Seven children (5 boys and 2 girls) with juxtacortical chondroma were identified, ranging in age from 6 years to 16 years (mean 12.3 years). Mild pain and a palpable mass were present in all seven children. Plain radiographs were available in 6/7, MR in 7/7, CT in 4/7 and skeletal scintigraphy in 5/7 children. Three lesions were located in the proximal humerus, with one each in the distal radius, distal femur, proximal tibia and scapula. Radiographic and CT features deemed highly suggestive of juxtacortical chondroma included cortical scalloping, underlying cortical sclerosis and overhanging margins. MRI features consistent with juxtacortical chondroma included isointensity to skeletal muscle on T1, marked hyperintensity on T2 and peripheral rim enhancement after contrast agent administration. One of seven lesions demonstrated intramedullary extension, and 2/7 showed adjacent soft-tissue edema. Juxtacortical chondroma is an uncommon benign lesion in children with characteristic features on plain radiographs, CT and MR. Recognition of these features is invaluable in guiding appropriate surgical management. (orig.)

  15. Mass-like extramedullary hematopoiesis: imaging features

    Ginzel, Andrew W. [Synergy Radiology Associates, Houston, TX (United States); Kransdorf, Mark J.; Peterson, Jeffrey J.; Garner, Hillary W. [Mayo Clinic, Department of Radiology, Jacksonville, FL (United States); Murphey, Mark D. [American Institute for Radiologic Pathology, Silver Spring, MD (United States)


    To report the imaging appearances of mass-like extramedullary hematopoiesis (EMH), to identify those features that are sufficiently characteristic to allow a confident diagnosis, and to recognize the clinical conditions associated with EMH and the relative incidence of mass-like disease. We retrospectively identified 44 patients with EMH; 12 of which (27%) had focal mass-like lesions and formed the study group. The study group consisted of 6 male and 6 female subjects with a mean age of 58 years (range 13-80 years). All 12 patients underwent CT imaging and 3 of the 12 patients had undergone additional MR imaging. The imaging characteristics of the extramedullary hematopoiesis lesions in the study group were analyzed and recorded. The patient's clinical presentation, including any condition associated with extramedullary hematopoiesis, was also recorded. Ten of the 12 (83%) patients had one or more masses located along the axial skeleton. Of the 10 patients with axial masses, 9 (90%) had multiple masses and 7 (70%) demonstrated internal fat. Eight patients (80%) had paraspinal masses and 4 patients (40%) had presacral masses. Seven patients (70%) had splenomegaly. Eleven of the 12 patients had a clinical history available for review. A predisposing condition for extramedullary hematopoiesis was present in 10 patients and included various anemias (5 cases; 45%), myelofibrosis/myelodysplastic syndrome (4 cases; 36%), and marrow proliferative disorder (1 case; 9%). One patient had no known predisposing condition. Mass-like extramedullary hematopoiesis most commonly presents as multiple, fat-containing lesions localized to the axial skeleton. When these imaging features are identified, extramedullary hematopoiesis should be strongly considered, particularly when occurring in the setting of a predisposing medical condition. (orig.)

  16. Erythrocyte Features for Malaria Parasite Detection in Microscopic Images of Thin Blood Smear: A Review

    Salam Shuleenda Devi


    Full Text Available Microscopic image analysis of blood smear plays a very important role in characterization of erythrocytes in screening of malaria parasites. The characteristics feature of erythrocyte changes due to malaria parasite infection. The microscopic features of the erythrocyte include morphology, intensity and texture. In this paper, the different features used to differentiate the non- infected and malaria infected erythrocyte have been reviewed.

  17. Imaging internal features of whole, unfixed bacteria.

    Thomson, Nicholas M; Channon, Kevin; Mokhtar, Noor Azlin; Staniewicz, Lech; Rai, Ranjana; Roy, Ipsita; Sato, Shun; Tsuge, Takeharu; Donald, Athene M; Summers, David; Sivaniah, Easan


    Wet scanning-transmission electron microscopy (STEM) is a technique that allows high-resolution transmission imaging of biological samples in a hydrated state, with minimal sample preparation. However, it has barely been used for the study of bacterial cells. In this study, we present an analysis of the advantages and disadvantages of wet STEM compared with standard transmission electron microscopy (TEM). To investigate the potential applications of wet STEM, we studied the growth of polyhydroxyalkanoate and triacylglycerol carbon storage inclusions. These were easily visible inside cells, even in the early stages of accumulation. Although TEM produces higher resolution images, wet STEM is useful when preservation of the sample is important or when studying the relative sizes of different features, since samples do not need to be sectioned. Furthermore, under carefully selected conditions, it may be possible to maintain cell viability, enabling new types of experiments to be carried out. To our knowledge, internal features of bacterial cells have not been imaged previously by this technique.

  18. Imaging features of foot osteoid osteoma

    Shukla, Satyen; Clarke, Andrew W.; Saifuddin, Asif [Royal National Orthopaedic Hospital NHS Trust, Department of Radiology, Stanmore, Middlesex (United Kingdom)


    We performed a retrospective review of the imaging of nine patients with a diagnosis of foot osteoid osteoma (OO). Radiographs, computed tomography (CT) and magnetic resonance imaging (MRI) had been performed in all patients. Radiographic features evaluated were the identification of a nidus and cortical thickening. CT features noted were nidus location (affected bone - intramedullary, intracortical, subarticular) and nidus calcification. MRI features noted were the presence of an identifiable nidus, presence and grade of bone oedema and whether a joint effusion was identified. Of the nine patients, three were female and six male, with a mean age of 21 years (range 11-39 years). Classical symptoms of OO (night pain, relief with aspirin) were identified in five of eight (62.5%) cases (in one case, the medical records could not be retrieved). In five patients the lesion was located in the hindfoot (four calcaneus, one talus), while four were in the mid- or forefoot (two metatarsal and two phalangeal). Radiographs were normal in all patients with hindfoot OO. CT identified the nidus in all cases (89%) except one terminal phalanx lesion, while MRI demonstrated a nidus in six of nine cases (67%). The nidus was of predominantly intermediate signal intensity on T1-weighted (T1W) sequences, with intermediate to high signal intensity on T2-weighted (T2W) sequences. High-grade bone marrow oedema, limited to the affected bone and adjacent soft tissue oedema was identified in all cases. In a young patient with chronic hindfoot pain and a normal radiograph, MRI features suggestive of possible OO include extensive bone marrow oedema limited to one bone, with a possible nidus demonstrated in two-thirds of cases. The presence or absence of a nidus should be confirmed with high-resolution CT. (orig.)

  19. Feature extraction & image processing for computer vision

    Nixon, Mark


    This book is an essential guide to the implementation of image processing and computer vision techniques, with tutorial introductions and sample code in Matlab. Algorithms are presented and fully explained to enable complete understanding of the methods and techniques demonstrated. As one reviewer noted, ""The main strength of the proposed book is the exemplar code of the algorithms."" Fully updated with the latest developments in feature extraction, including expanded tutorials and new techniques, this new edition contains extensive new material on Haar wavelets, Viola-Jones, bilateral filt

  20. 可逆性后部脑病综合征的临床、影像学特征及鉴别诊断分析%Clinical Features,Imaging Findings and Differential Diagnosis of Posterior Reversible Encephalopathy Syndrome

    李凤; 李琳琳; 陈秋; 喻明; 庞洪波


    目的:分析可逆性后部脑病综合征(PRES)的临床及影像学特征,提高临床对该病的诊断水平。方法回顾性分析遂宁市中心医院2010年1月—2014年7月收治的8例 PRES 患者的临床特征、脑脊液检查结果、脑电图检查结果、影像学检查结果以及患者治疗、预后情况,总结其鉴别诊断要点。结果8例患者均有头痛、视物不清症状,4例患者出现恶心呕吐及肢体抽搐,3例患者出现意识和精神障碍。基础疾病:妊娠 4例,高血压2例,肾上腺腺瘤1例,系统性红斑狼疮1例。2例患者颅脑 CTA 示双侧大脑中动脉、大脑后动脉及大脑前动脉血管分支节段性舒张或收缩。8例患者颅脑 MRI 示病变多数对称分布于双侧大脑后部白质区,部分皮质受累,T2WI 及 FLAIR 均呈高信号,边界不清。8例患者经相关治疗后神经系统症状消失,MRI 复查病灶大部分或完全消失。结论 PRES 病因及临床表现多样,病程具有可逆性,确诊时应结合影像学检查以排除颅内其他疾病。%Objective To analyze the clinical features and imaging findings of posterior reversible encephalopathy syndrome(PRES),to improve the diagnostic ability of clinicians. Methods From January 2010 to July 2014 in Suining Central Hospital,a total of 8 patients with PRES were enrolled,and their clinical features,cerebrospinal fluid examination results,electroencephalogram examination results,imaging findings,treatment and prognosis were retrospectively analyzed,the key points of differential diagnosis were summarized. Results All of the 8 cases occurred headache and blurred vision,4 cases occurred nausea and vomiting,limb seizures,3 cases occurred consciousness and mental disorders. Basic diseases:4 cases with pregnancy,2 cases with hypertension,1 case with adrenal adenoma,1 case with systemic lupus erythematosus. CTA findings:2 cases showed segmental relaxation or shrink of bilateral middle

  1. Cardiac tumors: CT and MR imaging features; Tumeurs cardiaques: aspects en scanner et en IRM

    Moskovitch, G.; Chabbert, V.; Escourrou, G.; Desloques, L.; Otal, P.; Glock, Y.; Rousseau, H. [Centre Hospitalier Universitaire de Rangueil, Service de Radiologie Generale, 31 - Toulouse (France)


    The CT and MR imaging features of the main cardiac tumors will be reviewed. Cross-sectional imaging features may help differentiate between cardiac tumors and pseudo-tumoral lesions and identify malignant features. Based on clinical features, imaging findings are helpful to further characterize the nature of the lesion. CT and MR imaging can demonstrate the relationship of the tumor with adjacent anatomical structures and are invaluable in the pre-surgical work-up and post-surgical follow-up. (authors)

  2. Feature selection with the image grand tour

    Marchette, David J.; Solka, Jeffrey L.


    The grand tour is a method for visualizing high dimensional data by presenting the user with a set of projections and the projected data. This idea was extended to multispectral images by viewing each pixel as a multidimensional value, and viewing the projections of the grand tour as an image. The user then looks for projections which provide a useful interpretation of the image, for example, separating targets from clutter. We discuss a modification of this which allows the user to select convolution kernels which provide useful discriminant ability, both in an unsupervised manner as in the image grand tour, or in a supervised manner using training data. This approach is extended to other window-based features. For example, one can define a generalization of the median filter as a linear combination of the order statistics within a window. Thus the median filter is that projection containing zeros everywhere except for the middle value, which contains a one. Using the convolution grand tour one can select projections on these order statistics to obtain new nonlinear filters.

  3. Unusual features of the sperm head differentiation in Mabuya quinquetaeniata.

    Ismail, M F


    Sperm head differentiation in Mabuya quinquetaeniata agrees in the main features with that of Agama adramitana, Agama blandfordi, and Acanthodactylus boskianus. The development of subacrosomal lateral canals, the disappearance of translucent medulla, and the existence of unilateral dense acrosome are new findings of the present investigation.

  4. 胰腺常见囊性肿瘤的影像表现特征及其鉴别诊断%Imaging features and differential diagnosis of common cystic tumors of pancreas

    朱跃强; 白人驹; 孙浩然; 李亚军


    Objective To investigate the imaging features and differential diagnosis points of common cystic tumors of the pancreas on MSCT and MRI. Methods All imaging data of 41 patients (10 serous cystic neoplasms [SEN], 14 mucinous cystic neoplasms [MCN], 17 intraductal papillary mucinous neoplasms [IPMN]) confirmed by surgery and pathology were retrospectively analyzed. Results Seven serous microcystadenomas manifested as honeycomb masses with small and numerous cysts, lobulated contour, fibrous central scar. Two serous oligocystadenomas (1 unilocular and 1 multilocular) showed as masses with large and less cysts, lobulated contour. One solid serous adenoma markedly enhanced on CT, while T2WI revealed the cystic features. MCN manifested as masses with large and less cysts and smooth contour. IPMN were communicated with pancreatic duct, showed clubbed fingerlike cysts in unilocular and pleomorphic in multilocular. IPMN showed proximal, distal or whole pancreatic duct dilatation, while SEN and MCN only showed proximal dilatation. Conclusion Different shapes of cysts are helpful to differential diagnosis of common cystic tumors of pancreas. Distal and whole pancreatic duct dilatation only occur in IPMN .%目的 探讨胰腺常见囊性肿瘤的MSCT和MRI表现特征及其鉴别诊断要点.方法 回顾性分析经手术病理证实的41例胰腺囊性肿瘤[浆液性囊性肿瘤(SCN)10例,黏液性囊性肿瘤(MCN)14例,导管内乳头状黏液性肿瘤(IPMN)17例]的MSCT和MRI表现.结果 SCN中浆液性微囊性腺瘤7例,呈分叶状,囊小而多,多具有中心瘢痕;浆液性寡囊性腺瘤2例,1例单房、1例多房,囊大而少,边缘分叶;实性浆液性腺瘤1例,CT增强检查明显强化,但T2WI可显示其囊性特征.MCN囊大而少,边缘多光滑.IPMN与胰管相通,单房者多表现为杵状指样的囊,多房者囊常呈多种形态.IPMN胰管可出现远端、近端或全程扩张,而SCN和MCN仅近端扩张.结论 胰腺常见囊性肿瘤中囊的不

  5. Aggressive primary thyroid lymphoma: imaging features of two elderly patients

    Eu Hyun Kim


    Full Text Available

    We report two cases of aggressive thyroid lymphoma in elderly patients that presented as Epub ahead of print large infiltrative thyroid masses with extensive invasion to adjacent structures including trachea, esophagus, and common carotid artery. Ultrasonography displayed irregular shaped, heterogeneous hypoechoic mass, mimicking anaplastic carcinoma. Computed tomography showed heterogeneously enhancing mass compared to surrounding muscles without calcification and hemorrhage. After biopsy, the masses were histopathologically diagnosed as lymphoma. Aggressive primary thyroid lymphoma is rare; therefore, here we report its image features, with emphasis on ultrasonographic findings, and discuss its differential diagnosis.

  6. Aggressive primary thyroid lymphoma: imaging features of two elderly patients

    Kim, Eu Hyun; Kim, Jee Young; Kim, Tae Jung [Yeouido St. Mary' s Hospital, The Catholic University College of Medicine, Seoul (Korea, Republic of)


    We report two cases of aggressive thyroid lymphoma in elderly patients that presented as Epub ahead of print large infiltrative thyroid masses with extensive invasion to adjacent structures including trachea, esophagus, and common carotid artery. Ultrasonography displayed irregular shaped, heterogeneous hypoechoic mass, mimicking anaplastic carcinoma. Computed tomography showed heterogeneously enhancing mass compared to surrounding muscles without calcification and hemorrhage. After biopsy, the masses were histopathologically diagnosed as lymphoma. Aggressive primary thyroid lymphoma is rare; therefore, here we report its image features, with emphasis on ultrasonographic findings, and discuss its differential diagnosis.

  7. Analysis of the features of contrast-enhanced ultrasound imaging of bacterial hepatic abscess and its value of differential diagnosis%细菌性肝脓肿患者超声造影特点及诊断分析

    史景璐; 曹青峰; 陈国勇


    目的:分析细菌性肝脓肿患者超声造影特点,以期提高诊断符合率。方法选择2009年1月-2013年12月经超声造影检查的30例细菌性肝脓肿患者41个病灶资料,回顾性分析细菌性肝脓肿常规超声与超声造影表现,比较两种检查方法的诊断符合率。结果30例细菌性肝脓肿患者中有21例单发、9例多发,累及部位肝右叶18例、肝左叶9例、肝左叶均累及3例,脓肿大小为直径2~9 cm ,平均(5.18±2.76)cm ;超声造影检查:41个病灶均表现病灶缩小征,其中17个病灶呈花瓣样强化、14个病灶呈环状强化、10个病灶呈蜂窝状强化;24例患者28个病灶呈典型的陡直快进快出曲线;超声造影诊断符合率为93.33%,明显高于常规超声的73.33%,差异有统计学意义(χ2=9.24,P<0.05)。结论超声造影检查集中了灰阶成像、彩色多普勒血流频谱和强化时间‐强度曲线分析的优势,对提高细菌性肝脓肿超声诊断符合率有较高的应用价值。%OBJECTIVE To analyze the features of contrast‐enhanced ultrasound imaging of bacterial hepatic abscess and its value of differential diagnosis so as to increase diagnostic accordance rate .METHODS Totally 30 patients with 41 bacterial liver abscess lesions were enrolled and they underwent ultrasound examinations during Jan .2009 to Dec .2013 .The conventional ultrasound and ultrasound imaging performance were retrospectively analyzed to compare the diagnostic accordance rate of the two methods of examination .RESULTS The 30 patients included 21 single lesions ,9 multiple lesions ,18 involving the right liver lobe ,9 on the left hepatic lobe ,and 3 involving both lobes ,with the abscess size of 2 -9 cm in diameter ,the average (5 .18 ± 2 .76) cm .The contrast‐enhanced ultrasound imaging showed 41 lesions narrowed ,including 17 lesions of strengthened petals ,14 lesions of cricoid reinforcement ,and 10

  8. Unsupervised feature learning for autonomous rock image classification

    Shu, Lei; McIsaac, Kenneth; Osinski, Gordon R.; Francis, Raymond


    Autonomous rock image classification can enhance the capability of robots for geological detection and enlarge the scientific returns, both in investigation on Earth and planetary surface exploration on Mars. Since rock textural images are usually inhomogeneous and manually hand-crafting features is not always reliable, we propose an unsupervised feature learning method to autonomously learn the feature representation for rock images. In our tests, rock image classification using the learned features shows that the learned features can outperform manually selected features. Self-taught learning is also proposed to learn the feature representation from a large database of unlabelled rock images of mixed class. The learned features can then be used repeatedly for classification of any subclass. This takes advantage of the large dataset of unlabelled rock images and learns a general feature representation for many kinds of rocks. We show experimental results supporting the feasibility of self-taught learning on rock images.

  9. Differential ghost imaging in time domain

    O-oka, Yoshiki; Fukatsu, Susumu


    Differential ghost imaging is attempted in time domain, i.e., temporal differential ghost imaging (TDGI), using pseudo-randomized light pulses and a temporal object consisting of no-return-to-zero bit patterns of varying duty. Evaluation of the signal-to-noise characteristics by taking into account errors due to false cross-correlation between the reference and the bucket detector readings indicates that the TDGI outperforms its non-differential counterpart, i.e., time-domain GI, in terms of consistently high and even duty-independent signal-to-noise ratios that are achieved.

  10. Local features in natural images via singularity theory

    Damon, James; Haslinger, Gareth


    This monograph considers a basic problem in the computer analysis of natural images, which are images of scenes involving multiple objects that are obtained by a camera lens or a viewer’s eye. The goal is to detect geometric features of objects in the image and to separate regions of the objects with distinct visual properties. When the scene is illuminated by a single principal light source, we further include the visual clues resulting from the interaction of the geometric features of objects, the shade/shadow regions on the objects, and the “apparent contours”. We do so by a mathematical analysis using a repertoire of methods in singularity theory. This is applied for generic light directions of both the “stable configurations” for these interactions, whose features remain unchanged under small viewer movement, and the generic changes which occur under changes of view directions. These may then be used to differentiate between objects and determine their shapes and positions.

  11. Hybridizing Differential Evolution with a Genetic Algorithm for Color Image Segmentation

    R. V. V. Krishna


    Full Text Available This paper proposes a hybrid of differential evolution and genetic algorithms to solve the color image segmentation problem. Clustering based color image segmentation algorithms segment an image by clustering the features of color and texture, thereby obtaining accurate prototype cluster centers. In the proposed algorithm, the color features are obtained using the homogeneity model. A new texture feature named Power Law Descriptor (PLD which is a modification of Weber Local Descriptor (WLD is proposed and further used as a texture feature for clustering. Genetic algorithms are competent in handling binary variables, while differential evolution on the other hand is more efficient in handling real parameters. The obtained texture feature is binary in nature and the color feature is a real value, which suits very well the hybrid cluster center optimization problem in image segmentation. Thus in the proposed algorithm, the optimum texture feature centers are evolved using genetic algorithms, whereas the optimum color feature centers are evolved using differential evolution.

  12. Iterative Reconstruction for Differential Phase Contrast Imaging

    Koehler, T.; Brendel, B.; Roessl, E.


    Purpose: The purpose of this work is to combine two areas of active research in tomographic x-ray imaging. The first one is the use of iterative reconstruction techniques. The second one is differential phase contrast imaging (DPCI). Method: We derive an SPS type maximum likelihood (ML) reconstructi

  13. Toward Automated Feature Detection in UAVSAR Images

    Parker, J. W.; Donnellan, A.; Glasscoe, M. T.


    Edge detection identifies seismic or aseismic fault motion, as demonstrated in repeat-pass inteferograms obtained by the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) program. But this identification is not robust at present: it requires a flattened background image, interpolation into missing data (holes) and outliers, and background noise that is either sufficiently small or roughly white Gaussian. Identification and mitigation of nongaussian background image noise is essential to creating a robust, automated system to search for such features. Clearly a robust method is needed for machine scanning of the thousands of UAVSAR repeat-pass interferograms for evidence of fault slip, landslides, and other local features.Empirical examination of detrended noise based on 20 km east-west profiles through desert terrain with little tectonic deformation for a suite of flight interferograms shows nongaussian characteristics. Statistical measurement of curvature with varying length scale (Allan variance) shows nearly white behavior (Allan variance slope with spatial distance from roughly -1.76 to -2) from 25 to 400 meters, deviations from -2 suggesting short-range differences (such as used in detecting edges) are often freer of noise than longer-range differences. At distances longer than 400 m the Allan variance flattens out without consistency from one interferogram to another. We attribute this additional noise afflicting difference estimates at longer distances to atmospheric water vapor and uncompensated aircraft motion.Paradoxically, California interferograms made with increasing time intervals before and after the El Mayor Cucapah earthquake (2008, M7.2, Mexico) show visually stronger and more interesting edges, but edge detection methods developed for the first year do not produce reliable results over the first two years, because longer time spans suffer reduced coherence in the interferogram. The changes over time are reflecting fault slip and block

  14. Imaging features of intraosseous ganglia: a report of 45 cases

    Williams, H.J.; Davies, A.M.; Allen, G.; Evans, N. [Royal Orthopaedic Hospital, Department of Radiology, Birmingham (United Kingdom); Mangham, D.C. [Royal Orthopaedic Hospital, Department of Pathology, Birmingham (United Kingdom)


    The aim of this study is to report the spectrum of imaging findings of intraosseous ganglia (IG) with particular emphasis on the radiographic and magnetic resonance (MR) features. Forty-five patients with a final diagnosis of IG were referred to a specialist orthopaedic oncology service with the presumptive diagnosis of either a primary or secondary bone tumour. The diagnosis was established by histology in 25 cases. In the remainder, the imaging features were considered characteristic and the lesion was stable on follow-up radiographic examination. Radiographs were available for retrospective review in all cases and MR imaging in 29. There was a minor male preponderance with a wide adult age range. Three quarters were found in relation to the weight-bearing long bones of the lower limb, particularly round the knee. On radiographs all were juxta-articular and osteolytic; 74% were eccentric in location, 80% had a sclerotic endosteal margin and 60% of cases showed a degree of trabeculation. Periosteal new bone formation and matrix mineralization were not present. Of the 29 cases that underwent MR imaging, 66% were multiloculated. On T1-weighted images the IG contents were isointense or mildly hypointense in 90% cases. Forty-one per cent of the cases showed a slightly hyperintense rim lining that enhanced with a gadolinium chelate. Thirty-eight per cent were associated with soft tissue extension and 17% with a defect of the adjacent articular cortex. Fifty-five per cent showed surrounding marrow oedema on T2-weighted or STIR images and two cases (7%) a fluid-fluid level prior to any surgical intervention. The authors contend that it is semantics to differentiate between an IG and a degenerate subchondral cyst as, while the initial pathogenesis may vary, the histological endpoint is identical, as are the imaging features apart from the degree of associated degenerative joint disease. IGs, particularly when large, may be mistaken for a bone tumour. Correlation of the

  15. Color and neighbor edge directional difference feature for image retrieval

    Chaobing Huang; Shengsheng Yu; Jingli Zhou; Hongwei Lu


    @@ A novel image feature termed neighbor edge directional difference unit histogram is proposed, in which the neighbor edge directional difference unit is defined and computed for every pixel in the image, and is used to generate the neighbor edge directional difference unit histogram. This histogram and color histogram are used as feature indexes to retrieve color image. The feature is invariant to image scaling and translation and has more powerful descriptive for the natural color images. Experimental results show that the feature can achieve better retrieval performance than other color-spatial features.

  16. Fast Fractal Image Encoding Based on Special Image Features

    ZHANG Chao; ZHOU Yiming; ZHANG Zengke


    The fractal image encoding method has received much attention for its many advantages over other methods,such as high decoding quality at high compression ratios. However, because every range block must be compared to all domain blocks in the codebook to find the best-matched one during the coding procedure, baseline fractal coding (BFC) is quite time consuming. To speed up fractal coding, a new fast fractal encoding algorithm is proposed. This algorithm aims at reducing the size of the search window during the domain-range matching process to minimize the computational cost. A new theorem presented in this paper shows that a special feature of the image can be used to do this work. Based on this theorem, the most inappropriate domain blocks, whose features are not similar to that of the given range block, are excluded before matching. Thus, the best-matched block can be captured much more quickly than in the BFC approachThe experimental results show that the runtime of the proposed method is reduced greatly compared to the BFC method. At the same time,the new algorithm also achieves high reconstructed image quality. In addition,the method can be incorporated with other fast algorithms to achieve better performance.Therefore, the proposed algorithm has a much better application potential than BFC.

  17. Accurate Image Retrieval Algorithm Based on Color and Texture Feature

    Chunlai Yan


    Full Text Available Content-Based Image Retrieval (CBIR is one of the most active hot spots in the current research field of multimedia retrieval. According to the description and extraction of visual content (feature of the image, CBIR aims to find images that contain specified content (feature in the image database. In this paper, several key technologies of CBIR, e. g. the extraction of the color and texture features of the image, as well as the similarity measures are investigated. On the basis of the theoretical research, an image retrieval system based on color and texture features is designed. In this system, the Weighted Color Feature based on HSV space is adopted as a color feature vector, four features of the Co-occurrence Matrix, saying Energy, Entropy, Inertia Quadrature and Correlation, are used to construct texture vectors, and the Euclidean distance for similarity measure is employed as well. Experimental results show that this CBIR system is efficient in image retrieval.

  18. Image retrieval using both color and texture features


    In order to improve the retrieval performance of images, this paper proposes an efficient approach for extracting and retrieving color images. The block diagram of our proposed approach to content-based image retrieval (CBIR) is given firstly, and then we introduce three image feature extracting arithmetic including color histogram, edge histogram and edge direction histogram, the histogram Euclidean distance, cosine distance and histogram intersection are used to measure the image level similarity. On the basis of using color and texture features separately, a new method for image retrieval using combined features is proposed. With the test for an image database including 766 general-purpose images and comparison and analysis of performance evaluation for features and similarity measures, our proposed retrieval approach demonstrates a promising performance. Experiment shows that combined features are superior to every single one of the three features in retrieval.

  19. Imaging systems and applications: introduction to the feature.

    Imai, Francisco H; Linne von Berg, Dale C; Skauli, Torbjørn; Tominaga, Shoji; Zalevsky, Zeev


    Imaging systems have numerous applications in industrial, military, consumer, and medical settings. Assembling a complete imaging system requires the integration of optics, sensing, image processing, and display rendering. This issue features original research ranging from design of stimuli for human perception, optics applications, and image enhancement to novel imaging modalities in both color and infrared spectral imaging, gigapixel imaging as well as a systems perspective to imaging.

  20. Backscattering Differential Ghost Imaging in Turbid Media

    Bina, M; Molteni, M; Gatti, A; Lugiato, L A; Ferri, F


    In this Letter we present experimental results concerning the retrieval of images of absorbing objects immersed in turbid media via differential ghost imaging (DGI) in a backscattering configuration. The method has been applied, for the first time to our knowledge, to the imaging of small thin black objects located at different depths inside a turbid solution of polystyrene nanospheres and its performances assessed via comparison with standard imaging techniques. A simple theoretical model capable of describing the basic optics of DGI in turbid media is proposed.

  1. Image feature detectors and descriptors foundations and applications

    Hassaballah, Mahmoud


    This book provides readers with a selection of high-quality chapters that cover both theoretical concepts and practical applications of image feature detectors and descriptors. It serves as reference for researchers and practitioners by featuring survey chapters and research contributions on image feature detectors and descriptors. Additionally, it emphasizes several keywords in both theoretical and practical aspects of image feature extraction. The keywords include acceleration of feature detection and extraction, hardware implantations, image segmentation, evolutionary algorithm, ordinal measures, as well as visual speech recognition. .

  2. Mueller matrix polarimetry for differentiating characteristic features of cancerous tissues.

    Du, E; He, Honghui; Zeng, Nan; Sun, Minghao; Guo, Yihong; Wu, Jian; Liu, Shaoxiong; Ma, Hui


    Polarization measurements allow one to enhance the imaging contrast of superficial tissues and obtain new polarization sensitive parameters for better descriptions of the micro- and macro- structural and optical properties of complex tissues. Since the majority of cancers originate in the epithelial layer, probing the morphological and pathological changes in the superficial tissues using an expended parameter set with improved contrast will assist in early clinical detection of cancers. We carry out Mueller matrix imaging on different cancerous tissues to look for cancer specific features. Using proper scattering models and Monte Carlo simulations, we examine the relationship between the microstructures of the samples, which are represented by the parameters of the scattering model and the characteristic features of the Mueller matrix. This study gives new clues on the contrast mechanisms of polarization sensitive measurements for different cancers and may provide new diagnostic techniques for clinical applications.

  3. Can magnetic resonance imaging differentiate undifferentiated arthritis?

    Østergaard, Mikkel; Duer, Anne; Hørslev-Petersen, K


    A high sensitivity for the detection of inflammatory and destructive changes in inflammatory joint diseases makes magnetic resonance imaging potentially useful for assigning specific diagnoses, such as rheumatoid arthritis and psoriatic arthritis in arthritides, that remain undifferentiated after...... conventional clinical, biochemical and radiographic examinations. With recent data as the starting point, the present paper describes the current knowledge on magnetic resonance imaging in the differential diagnosis of undifferentiated arthritis....

  4. Comparative Study of Triangulation based and Feature based Image Morphing

    Ms. Bhumika G. Bhatt


    Full Text Available Image Morphing is one of the most powerful Digital Image processing technique, which is used to enhancemany multimedia projects, presentations, education and computer based training. It is also used inmedical imaging field to recover features not visible in images by establishing correspondence of featuresamong successive pair of scanned images. This paper discuss what morphing is and implementation ofTriangulation based morphing Technique and Feature based Image Morphing. IT analyze both morphingtechniques in terms of different attributes such as computational complexity, Visual quality of morphobtained and complexity involved in selection of features.

  5. Fingerprint image enhancement by differential hysteresis processing.

    Blotta, Eduardo; Moler, Emilce


    A new method to enhance defective fingerprints images through image digital processing tools is presented in this work. When the fingerprints have been taken without any care, blurred and in some cases mostly illegible, as in the case presented here, their classification and comparison becomes nearly impossible. A combination of spatial domain filters, including a technique called differential hysteresis processing (DHP), is applied to improve these kind of images. This set of filtering methods proved to be satisfactory in a wide range of cases by uncovering hidden details that helped to identify persons. Dactyloscopy experts from Policia Federal Argentina and the EAAF have validated these results.

  6. Cutaneous syncytial myoepithelioma:: Clinico-pathological features and differential diagnosis.

    Pizzi, Marco; Facchin, Federico; Kohlscheen, Eva; Sartore, Leonardo; Salmaso, Roberto; Bassetto, Franco


    Cutaneous syncytial myoepithelioma (CSM) is a very rare tumor belonging to the spectrum of skin myoepithelial lesions. CSM usually affects the upper extremities of young to middle aged patients and is characterized by peculiar morphological and immunohistochemical features. Unlike classic myoepithelioma, CSM is composed by a densely packed proliferation of spindled to histiocytoid cells, which are variably positive for EMA, S100, SMA, and frequently negative for cytokeratins and GFAP. The peculiar histopathology and the extreme rarity of such lesion (less than 40 cases reported in the literature) can make the diagnosis of CSM a true challenge. In the present case, we report the clinico-pathological features of a primary CSM occurring in a 38 year-old Caucasian man. The differential diagnoses of such lesion are also briefly discussed.

  7. Feature coding for image representation and recognition

    Huang, Yongzhen


    This brief presents a comprehensive introduction to feature coding, which serves as a key module for the typical object recognition pipeline. The text offers a rich blend of theory and practice while reflects the recent developments on feature coding, covering the following five aspects: (1) Review the state-of-the-art, analyzing the motivations and mathematical representations of various feature coding methods; (2) Explore how various feature coding algorithms evolve along years; (3) Summarize the main characteristics of typical feature coding algorithms and categorize them accordingly; (4) D

  8. Tracking image features with PCA-SURF descriptors

    Pancham, A


    Full Text Available The tracking of moving points in image sequences requires unique features that can be easily distinguished. However, traditional feature descriptors are of high dimension, leading to larger storage requirement and slower computation. In this paper...

  9. Evaluation of textural features for multispectral images

    Bayram, Ulya; Can, Gulcan; Duzgun, Sebnem; Yalabik, Nese


    Remote sensing is a field that has wide use, leading to the fact that it has a great importance. Therefore performance of selected features plays a great role. In order to gain some perspective on useful textural features, we have brought together state-of-art textural features in recent literature, yet to be applied in remote sensing field, as well as presenting a comparison with traditional ones. Therefore we selected most commonly used textural features in remote sensing that are grey-level co-occurrence matrix (GLCM) and Gabor features. Other selected features are local binary patterns (LBP), edge orientation features extracted after applying steerable filter, and histogram of oriented gradients (HOG) features. Color histogram feature is also used and compared. Since most of these features are histogram-based, we have compared performance of bin-by-bin comparison with a histogram comparison method named as diffusion distance method. During obtaining performance of each feature, k-nearest neighbor classification method (k-NN) is applied.

  10. New synthesizing feature parameter of wear particles image


    This paper outlines the application of wavelet analysis method to computering wear par-ticles image processing and introduces the concept of grain parameter for wear particle imagebased on statistical feature parameters. The feature of wear particles image can be obtained fromthe wavelet decomposition and the statistics analysis. Test results showed that grain parametercan be used as a synthesizing feature parameter for wear particle image.



    A new region feature which emphasized the salience of target region and its neighbors is proposed.In region segmentation-based multisensor image fusion scheme, the presented feature can be extracted from each segmented region to determine the fusion weight. Experimental results demonstrate that the proposed feature has extensive application scope and it provides much more information for each region. It can not only be used in image fusion but also be used in other image processing applications.

  12. Feature representation of RGB-D images using joint spatial-depth feature pooling

    Pan, Hong; Olsen, Søren Ingvor; Zhu, Yaping


    utilizes depth information only to extract local features, without considering it to improve robustness and discriminability of the feature representation by merging depth cues into feature pooling. Spatial pyramid model (SPM) has become the standard protocol to split a 2D image plane into sub......-regions for feature pooling of RGB-D images. We argue that SPM may not be the optimal pooling scheme for RGB-D images, as it only pools features spatially and completely discards their depth topological structures. Instead, we propose a novel joint spatial-depth pooling (JSDP) scheme which further partitions SPM...

  13. Imaging features of ovarian metastases from colonic adenocarcinoma in adolescents

    Kauffman, W.M. [Dept. of Diagnostic Imaging, St. Jude Children`s Research Hospital, Memphis, TN (United States); Jenkins, J.J. III [Dept. of Pathology, St. Jude Children`s Research Hospital, Memphis, TN (United States); Helton, K. [Dept. of Diagnostic Imaging, Vanderbilt Univ. Medical Center, Nashville, TN (United States); Rao, B.N. [Dept. of Surgery, St. Jude Children`s Research Hospital, Memphis, TN (United States); Winer-Muram, H.T. [Dept. of Diagnostic Imaging, St. Jude Children`s Research Hospital, Memphis, TN (United States); Pratt, C.B. [Dept. of Hematology-Oncology, St. Jude Children`s Research Hospital, Memphis, TN (United States)


    This paper describes the imaging features of ovarian metastases from adenocarcinoma of the colon in adolescent females. We reviewed retrospectively abdominal and pelvic computed tomographic and pelvic ultrasound examinations, histologic slices, and clinical charts of six adolescent females with ovarian metastases secondary to adenocarcinoma of the colon. One patient had ovarian metastasis at presentation and was presumed to have a primary ovarian tumor. The ovarian metastases were either solid (n = 3), complex with both solid and cystic components (n = 2), or multilocular cysts (n = 1). The ovarian lesions were large, ranging from 6 cm to 18 cm in diameter. Colorectal carcinoma in adolescent females is frequently associated with ovarian metastases. One imaging characteristic differs in adult and adolescent ovarian metastases, although they do have features in common: in adolescents, a smaller proportion of colorectal ovarian metastases are multicystic (17%) compared with the adult series (45%). These lesions are frequently large and may be complex, multicystic, or solid. Although it is a rare disease, the differential dignosis of adnexal masses in adolescent females should include ovarian metastases from adenocarcinoma of the colon. (orig.)

  14. Glioblastoma multiforme: exploratory radiogenomic analysis by using quantitative image features.

    Gevaert, Olivier; Mitchell, Lex A; Achrol, Achal S; Xu, Jiajing; Echegaray, Sebastian; Steinberg, Gary K; Cheshier, Samuel H; Napel, Sandy; Zaharchuk, Greg; Plevritis, Sylvia K


    To derive quantitative image features from magnetic resonance (MR) images that characterize the radiographic phenotype of glioblastoma multiforme (GBM) lesions and to create radiogenomic maps associating these features with various molecular data. Clinical, molecular, and MR imaging data for GBMs in 55 patients were obtained from the Cancer Genome Atlas and the Cancer Imaging Archive after local ethics committee and institutional review board approval. Regions of interest (ROIs) corresponding to enhancing necrotic portions of tumor and peritumoral edema were drawn, and quantitative image features were derived from these ROIs. Robust quantitative image features were defined on the basis of an intraclass correlation coefficient of 0.6 for a digital algorithmic modification and a test-retest analysis. The robust features were visualized by using hierarchic clustering and were correlated with survival by using Cox proportional hazards modeling. Next, these robust image features were correlated with manual radiologist annotations from the Visually Accessible Rembrandt Images (VASARI) feature set and GBM molecular subgroups by using nonparametric statistical tests. A bioinformatic algorithm was used to create gene expression modules, defined as a set of coexpressed genes together with a multivariate model of cancer driver genes predictive of the module's expression pattern. Modules were correlated with robust image features by using the Spearman correlation test to create radiogenomic maps and to link robust image features with molecular pathways. Eighteen image features passed the robustness analysis and were further analyzed for the three types of ROIs, for a total of 54 image features. Three enhancement features were significantly correlated with survival, 77 significant correlations were found between robust quantitative features and the VASARI feature set, and seven image features were correlated with molecular subgroups (P < .05 for all). A radiogenomics map was

  15. Intrinsic feature-based pose measurement for imaging motion compensation

    Baba, Justin S.; Goddard, Jr., James Samuel


    Systems and methods for generating motion corrected tomographic images are provided. A method includes obtaining first images of a region of interest (ROI) to be imaged and associated with a first time, where the first images are associated with different positions and orientations with respect to the ROI. The method also includes defining an active region in the each of the first images and selecting intrinsic features in each of the first images based on the active region. Second, identifying a portion of the intrinsic features temporally and spatially matching intrinsic features in corresponding ones of second images of the ROI associated with a second time prior to the first time and computing three-dimensional (3D) coordinates for the portion of the intrinsic features. Finally, the method includes computing a relative pose for the first images based on the 3D coordinates.

  16. Retinal image analysis: preprocessing and feature extraction

    Marrugo, Andres G; Millan, Maria S, E-mail: [Grup d' Optica Aplicada i Processament d' Imatge, Departament d' Optica i Optometria Univesitat Politecnica de Catalunya (Spain)


    Image processing, analysis and computer vision techniques are found today in all fields of medical science. These techniques are especially relevant to modern ophthalmology, a field heavily dependent on visual data. Retinal images are widely used for diagnostic purposes by ophthalmologists. However, these images often need visual enhancement prior to apply a digital analysis for pathological risk or damage detection. In this work we propose the use of an image enhancement technique for the compensation of non-uniform contrast and luminosity distribution in retinal images. We also explore optic nerve head segmentation by means of color mathematical morphology and the use of active contours.

  17. Image mosaic method based on SIFT features of line segment.

    Zhu, Jun; Ren, Mingwu


    This paper proposes a novel image mosaic method based on SIFT (Scale Invariant Feature Transform) feature of line segment, aiming to resolve incident scaling, rotation, changes in lighting condition, and so on between two images in the panoramic image mosaic process. This method firstly uses Harris corner detection operator to detect key points. Secondly, it constructs directed line segments, describes them with SIFT feature, and matches those directed segments to acquire rough point matching. Finally, Ransac method is used to eliminate wrong pairs in order to accomplish image mosaic. The results from experiment based on four pairs of images show that our method has strong robustness for resolution, lighting, rotation, and scaling.

  18. Introduction: feature issue on In Vivo Microcirculation Imaging.

    Dunn, Andrew K; Leitgeb, Rainer; Wang, Ruikang K; Zhang, Hao F


    The editors introduce the Biomedical Optics Express feature issue, "In Vivo Microcirculation Imaging," which includes 14 contributions from the biomedical optics community, covering such imaging techniques as optical coherence tomography, photoacoustic microscopy, laser Doppler /speckle imaging, and near infrared spectroscopy and fluorescence imaging.

  19. [Multiple transmission electron microscopic image stitching based on sift features].

    Li, Mu; Lu, Yanmeng; Han, Shuaihu; Wu, Zhuobin; Chen, Jiajing; Liu, Zhexing; Cao, Lei


    We proposed a new stitching method based on sift features to obtain an enlarged view of transmission electron microscopic (TEM) images with a high resolution. The sift features were extracted from the images, which were then combined with fitted polynomial correction field to correct the images, followed by image alignment based on the sift features. The image seams at the junction were finally removed by Poisson image editing to achieve seamless stitching, which was validated on 60 local glomerular TEM images with an image alignment error of 62.5 to 187.5 nm. Compared with 3 other stitching methods, the proposed method could effectively reduce image deformation and avoid artifacts to facilitate renal biopsy pathological diagnosis.

  20. Cystic synovial sarcomas: imaging features with clinical and histopathologic correlation

    Nakanishi, Hirofumi; Araki, Nobuhito [Department of Orthopedic Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3, Nakamichi, Higashinari-Ku, 537-8511, Osaka (Japan); Sawai, Yuka [Department of Radiology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka (Japan); Kudawara, Ikuo [Department of Orthopedic Surgery, Osaka National Hospital, Osaka (Japan); Mano, Masayuki; Ishiguro, Shingo [Department of Pathology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka (Japan); Ueda, Takafumi; Yoshikawa, Hideki [Department of Orthopedic Surgery, Osaka University Graduate School of Medicine, Suita, Osaka (Japan)


    To characterize the radiological and clinicopathologic features of cystic synovial sarcoma. Seven patients with primary cystic synovial sarcoma were evaluated. Computed tomography (CT) and magnetic resonance (MR) imaging were undertaken at the first presentation. The diagnosis of synovial sarcoma was made on the basis of histological examinations followed by molecular analysis. Radiological and clinicopathologic findings were reviewed. CT showed well-defined soft tissue mass without cortical bone erosion and invasion. Calcification was seen at the periphery of the mass in three cases. T2-weighted MR images showed multilocular inhomogeneous intensity mass in all cases, five of which showed fluid-fluid levels. On gross appearance, old and/or fresh hematomas were detected in six cases. In the one remaining case, microscopic hemorrhage in the cystic lumen was proven. Four cases had poorly differentiated areas. In five cases prominent hemangiopericytomatous vasculature was observed. Histologic grade was intermediate in one tumor and high in six. One case had a history of misdiagnosis for tarsal tunnel syndrome, one for lymphadenopathy, two for sciatica and two for hematoma. All cystic synovial sarcomas demonstrated multilocularity with well-circumscribed walls and internal septae. Synovial sarcoma should be taken into consideration in patients with deeply situated multicystic mass with triple signal intensity on T2-weighted MR imaging. (orig.)

  1. Remote Sensing Image Feature Extracting Based Multiple Ant Colonies Cooperation

    Zhang Zhi-long


    Full Text Available This paper presents a novel feature extraction method for remote sensing imagery based on the cooperation of multiple ant colonies. First, multiresolution expression of the input remote sensing imagery is created, and two different ant colonies are spread on different resolution images. The ant colony in the low-resolution image uses phase congruency as the inspiration information, whereas that in the high-resolution image uses gradient magnitude. The two ant colonies cooperate to detect features in the image by sharing the same pheromone matrix. Finally, the image features are extracted on the basis of the pheromone matrix threshold. Because a substantial amount of information in the input image is used as inspiration information of the ant colonies, the proposed method shows higher intelligence and acquires more complete and meaningful image features than those of other simple edge detectors.

  2. Modern imaging: introduction to the feature issue.

    Catrysse, Peter B; Irsch, Kristina; Javidi, Bahram; Preza, Chrysanthe; Testorf, Markus; Zalevsky, Zeev


    This special issue of Applied Optics contains selected papers reflecting the various disciplines that are needed for the design, implementation and advancement of imaging technology and systems, and it highlights the state-of-the-art research developments in the areas of modern imaging use.

  3. Load-differential features for automated detection of fatigue cracks using guided waves

    Chen, Xin; Lee, Sang Jun; Michaels, Jennifer E.; Michaels, Thomas E.


    Guided wave structural health monitoring (SHM) is being considered to assess the integrity of plate-like structures for many applications. Prior research has investigated how guided wave propagation is affected by applied loads, which induce anisotropic changes in both dimensions and phase velocity. In addition, it is well-known that applied tensile loads open fatigue cracks and thus enhance their detectability using ultrasonic methods. Here we describe load-differential methods in which signals recorded from different loads at the same damage state are compared without using previously obtained damage-free data. Changes in delay-and-sum images are considered as a function of differential loads and damage state. Load-differential features are extracted from these images that capture the effects of loading as fatigue cracks are opened. Damage detection thresholds are adaptively set based upon the load-differential behavior of the various features, which enables implementation of an automated fatigue crack detection process. The efficacy of the proposed approach is examined using data from a fatigue test performed on an aluminum plate specimen that is instrumented with a sparse array of surface-mounted ultrasonic guided wave transducers.

  4. A combinatorial Bayesian and Dirichlet model for prostate MR image segmentation using probabilistic image features

    Li, Ang; Li, Changyang; Wang, Xiuying; Eberl, Stefan; Feng, Dagan; Fulham, Michael


    Blurred boundaries and heterogeneous intensities make accurate prostate MR image segmentation problematic. To improve prostate MR image segmentation we suggest an approach that includes: (a) an image patch division method to partition the prostate into homogeneous segments for feature extraction; (b) an image feature formulation and classification method, using the relevance vector machine, to provide probabilistic prior knowledge for graph energy construction; (c) a graph energy formulation scheme with Bayesian priors and Dirichlet graph energy and (d) a non-iterative graph energy minimization scheme, based on matrix differentiation, to perform the probabilistic pixel membership optimization. The segmentation output was obtained by assigning pixels with foreground and background labels based on derived membership probabilities. We evaluated our approach on the PROMISE-12 dataset with 50 prostate MR image volumes. Our approach achieved a mean dice similarity coefficient (DSC) of 0.90  ±  0.02, which surpassed the five best prior-based methods in the PROMISE-12 segmentation challenge.

  5. Registration of multitemporal aerial optical images using line features

    Zhao, Chenyang; Goshtasby, A. Ardeshir


    Registration of multitemporal images is generally considered difficult because scene changes can occur between the times the images are obtained. Since the changes are mostly radiometric in nature, features are needed that are insensitive to radiometric differences between the images. Lines are geometric features that represent straight edges of rigid man-made structures. Because such structures rarely change over time, lines represent stable geometric features that can be used to register multitemporal remote sensing images. An algorithm to establish correspondence between lines in two images of a planar scene is introduced and formulas to relate the parameters of a homography transformation to the parameters of corresponding lines in images are derived. Results of the proposed image registration on various multitemporal images are presented and discussed.

  6. Remote sensing image classification based on block feature point density analysis and multiple-feature fusion

    Li, Shijin; Jiang, Yaping; Zhang, Yang; Feng, Jun


    With the development of remote sensing (RS) and the related technologies, the resolution of RS images is enhancing. Compared with moderate or low resolution images, high-resolution ones can provide more detailed ground information. However, a variety of terrain has complex spatial distribution. The different objectives of high-resolution images have a variety of features. The effectiveness of these features is not the same, but some of them are complementary. Considering the above information and characteristics, a new method is proposed to classify RS images based on hierarchical fusion of multi-features. Firstly, RS images are pre-classified into two categories in terms of whether feature points are uniformly or non-uniformly distributed. Then, the color histogram and Gabor texture feature are extracted from the uniformly-distributed categories, and the linear spatial pyramid matching using sparse coding (ScSPM) feature is obtained from the non-uniformly-distributed categories. Finally, the classification is performed by two support vector machine classifiers. The experimental results on a large RS image database with 2100 images show that the overall classification accuracy is boosted by 10.1% in comparison with the highest accuracy of single feature classification method. Compared with other multiple-feature fusion methods, the proposed method has achieved the highest classification accuracy on this dataset which has reached 90.1%, and the time complexity of the algorithm is also greatly reduced.

  7. Mining Mid-level Features for Image Classification

    Fernando, Basura; Fromont, Elisa; Tuytelaars, Tinne


    International audience; Mid-level or semi-local features learnt using class-level information are potentially more distinctive than the traditional low-level local features constructed in a purely bottom-up fashion. At the same time they preserve some of the robustness properties with respect to occlusions and image clutter. In this paper we propose a new and effective scheme for extracting mid-level features for image classification, based on relevant pattern mining. In par- ticular, we mine...

  8. Featured Image: A New Look at Fomalhaut

    Kohler, Susanna


    ALMA continuum image overlaid as contours on the Hubble STIS image of Fomalhaut. [MacGregor et al. 2017]This stunning image of the Fomalhaut star system was taken by the Atacama Large Millimeter/submillimeter Array (ALMA) in Chile. This image maps the 1.3-mm continuum emission from the dust around the central star, revealing a ring that marks the outer edge of the planet-forming debris disk surrounding the star. In a new study, a team of scientists led by Meredith MacGregor (Harvard-Smithsonian Center for Astrophysics) examines these ALMA observations of Fomalhaut, which beautifully complement former Hubble images of the system. ALMAs images provide the first robust detection of apocenter glow the brightening of the ring at the point farthest away from the central star, a side effect of the rings large eccentricity. The authors use ALMAsobservations to measure properties of the disk, such as its span (roughly 136 x 14 AU), eccentricity (e 0.12), and inclination angle ( 66). They then explore the implications for Fomalhaut b, the planet located near the outer disk. To read more about the teams observations, check out the paper below.CitationMeredith A. MacGregor et al 2017 ApJ 842 8. doi:10.3847/1538-4357/aa71ae

  9. Feature-based Alignment of Volumetric Multi-modal Images

    Toews, Matthew; Zöllei, Lilla; Wells, William M.


    This paper proposes a method for aligning image volumes acquired from different imaging modalities (e.g. MR, CT) based on 3D scale-invariant image features. A novel method for encoding invariant feature geometry and appearance is developed, based on the assumption of locally linear intensity relationships, providing a solution to poor repeatability of feature detection in different image modalities. The encoding method is incorporated into a probabilistic feature-based model for multi-modal image alignment. The model parameters are estimated via a group-wise alignment algorithm, that iteratively alternates between estimating a feature-based model from feature data, then realigning feature data to the model, converging to a stable alignment solution with few pre-processing or pre-alignment requirements. The resulting model can be used to align multi-modal image data with the benefits of invariant feature correspondence: globally optimal solutions, high efficiency and low memory usage. The method is tested on the difficult RIRE data set of CT, T1, T2, PD and MP-RAGE brain images of subjects exhibiting significant inter-subject variability due to pathology. PMID:24683955

  10. A Method for Image Decontamination Based on Partial Differential Equation

    Hou Junping


    Full Text Available This paper will introduce the method to apply partial differential equations for the decontamination processing of images. It will establish continuous partial differential mathematical models for image information and use specific solving methods to conduct decontamination processing to images during the process of solving partial differential equations, such as image noise reduction, image denoising and image segmentation. This paper will study the uniqueness of solution for the partial differential equations and the monotonicity that functional constrain has on multipliers by making analysis of the ROF model in the partial differential mathematical model.

  11. Feature extraction for an image retrieving scheme

    Fuertes García, José Manuel; Lucena López, Manuel José; Pérez de la Blanca Capilla, Nicolás; Fernández Valdivia, Joaquín


    In this paper we present two basic modules whithin the designed scheme for retrieving images of a database from the object colour and shape in the scenes. On the one hand, we desing a new method to detect edges in colour images. We offer a new approach to the perceptual space (H,S,I) (an Uniform Chromatic Scale space) About wich we describe its properties as well as the metric to work in it. On the other hand, we develop an information simplifying process to form a graphic structure en wich t...

  12. Segmentation-Based PolSAR Image Classification Using Visual Features: RHLBP and Color Features

    Jian Cheng


    Full Text Available A segmentation-based fully-polarimetric synthetic aperture radar (PolSAR image classification method that incorporates texture features and color features is designed and implemented. This method is based on the framework that conjunctively uses statistical region merging (SRM for segmentation and support vector machine (SVM for classification. In the segmentation step, we propose an improved local binary pattern (LBP operator named the regional homogeneity local binary pattern (RHLBP to guarantee the regional homogeneity in PolSAR images. In the classification step, the color features extracted from false color images are applied to improve the classification accuracy. The RHLBP operator and color features can provide discriminative information to separate those pixels and regions with similar polarimetric features, which are from different classes. Extensive experimental comparison results with conventional methods on L-band PolSAR data demonstrate the effectiveness of our proposed method for PolSAR image classification.

  13. Angular Differential Imaging: a Powerful High-Contrast Imaging Technique

    Marois, C; Lafreniere, D; Doyon, R; Macintosh, B; Nadeau, D


    Angular differential imaging is a high-contrast imaging technique that reduces speckle noise from quasi-static optical aberrations and facilitates the detection of faint nearby companions. A sequence of images is acquired with an altitude/azimuth telescope, the instrument rotator being turned off. This keeps the instrument and telescope optics aligned, stabilizes the instrumental PSF and allows the field of view to rotate with respect to the instrument. For each image, a reference PSF obtained from other images of the sequence is subtracted. All residual images are then rotated to align the field and are median combined. Observed performances are reported for Gemini Altair/NIRI data. Inside the speckle dominated region of the PSF, it is shown that quasi-static PSF noise can be reduced by a factor {approx}5 for each image subtraction. The combination of all residuals then provides an additional gain of the order of the square root of the total number of images acquired. To our knowledge, this is the first time an acquisition strategy and reduction pipeline designed for speckle attenuation and high contrast imaging is demonstrated to significantly get better detection limits with longer integration times at all angular separations. A PSF noise attenuation of 100 was achieved from 2-hour long sequences of images of Vega, reaching a 5-sigma contrast of 20 magnitudes for separations greater than 7''. This technique can be used with currently available instruments to search for {approx} 1 M{sub Jup} exoplanets with orbits of radii between 50 and 300 AU around nearby young stars. The possibility of combining the technique with other high-contrast imaging methods is briefly discussed.

  14. Feature extraction for magnetic domain images of magneto-optical recording films using gradient feature segmentation

    Quanqing, Zhu; Xinsai, Wang; Xuecheng, Zou; Haihua, Li; Xiaofei, Yang


    In this paper, we present a method to realize feature extraction on low contrast magnetic domain images of magneto-optical recording films. The method is based on the following three steps: first, Lee-filtering method is adopted to realize pre-filtering and noise reduction; this is followed by gradient feature segmentation, which separates the object area from the background area; finally the common linking method is adopted and the characteristic parameters of magnetic domain are calculated. We describe these steps with particular emphasis on the gradient feature segmentation. The results show that this method has advantages over other traditional ones for feature extraction of low contrast images.

  15. Registration of Standardized Histological Images in Feature Space

    Bagci, Ulas; 10.1117/12.770219


    In this paper, we propose three novel and important methods for the registration of histological images for 3D reconstruction. First, possible intensity variations and nonstandardness in images are corrected by an intensity standardization process which maps the image scale into a standard scale where the similar intensities correspond to similar tissues meaning. Second, 2D histological images are mapped into a feature space where continuous variables are used as high confidence image features for accurate registration. Third, we propose an automatic best reference slice selection algorithm that improves reconstruction quality based on both image entropy and mean square error of the registration process. We demonstrate that the choice of reference slice has a significant impact on registration error, standardization, feature space and entropy information. After 2D histological slices are registered through an affine transformation with respect to an automatically chosen reference, the 3D volume is reconstruct...

  16. Robust Image Hashing Using Radon Transform and Invariant Features

    Y.L. Liu


    Full Text Available A robust image hashing method based on radon transform and invariant features is proposed for image authentication, image retrieval, and image detection. Specifically, an input image is firstly converted into a counterpart with a normalized size. Then the invariant centroid algorithm is applied to obtain the invariant feature point and the surrounding circular area, and the radon transform is employed to acquire the mapping coefficient matrix of the area. Finally, the hashing sequence is generated by combining the feature vectors and the invariant moments calculated from the coefficient matrix. Experimental results show that this method not only can resist against the normal image processing operations, but also some geometric distortions. Comparisons of receiver operating characteristic (ROC curve indicate that the proposed method outperforms some existing methods in classification between perceptual robustness and discrimination.

  17. Featured Image: Active Cryovolcanism on Europa?

    Kohler, Susanna


    Nighttime thermal image from the Galileo Photopolarimeter-Radiometer, revealing a thermal anomaly around the region where the plumes were observed. [Sparks et al. 2017]This image shows a 1320 900 km, high-resolution Galileo/Voyager USGS map of the surface of Europa, one of Jupiters moons. In March 2014, observations of Europa revealed a plume on its icy surface coming from somewhere within the green ellipse. In February 2016, another plume was observed, this time originating from somewhere within the cyan ellipse. In addition, a nighttime thermal image from the Galileo Photopolarimeter-Radiometer has revealed a thermal anomaly a region of unusually high temperature near the same location. In a recent study led by William Sparks (Space Telescope Science Institute), a team of scientists presents these observations and argues that they provide mounting evidence of active water-vapor venting from ongoing cryovolcanism beneath Europas icy surface. If this is true, then Europas surface is active and provides access to the liquid water at depth boosting the case for Europas potential habitability and certainly making for an interesting target point for future spacecraft exploration of this moon. For more information, check out the paper below!CitationW. B. Sparks et al 2017 ApJL 839 L18. doi:10.3847/2041-8213/aa67f8

  18. Simple Low Level Features for Image Analysis

    Falcoz, Paolo

    As human beings, we perceive the world around us mainly through our eyes, and give what we see the status of “reality”; as such we historically tried to create ways of recording this reality so we could augment or extend our memory. From early attempts in photography like the image produced in 1826 by the French inventor Nicéphore Niépce (Figure 2.1) to the latest high definition camcorders, the number of recorded pieces of reality increased exponentially, posing the problem of managing all that information. Most of the raw video material produced today has lost its memory augmentation function, as it will hardly ever be viewed by any human; pervasive CCTVs are an example. They generate an enormous amount of data each day, but there is not enough “human processing power” to view them. Therefore the need for effective automatic image analysis tools is great, and a lot effort has been put in it, both from the academia and the industry. In this chapter, a review of some of the most important image analysis tools are presented.

  19. Featured Image: A Filament Forms and Erupts

    Kohler, Susanna


    This dynamic image of active region NOAA 12241 was captured by the Solar Dynamics Observatorys Atmospheric Imaging Assembly in December 2014. Observations of this region from a number of observatories and instruments recently presented by Jincheng Wang (University of Chinese Academy of Sciences) and collaborators reveal details about the formation and eruption of a long solar filament. Wang and collaborators show that the right part of the filament formed by magnetic reconnection between two bundles of magnetic field lines, while the left part formed as a result of shearing motion. When these two parts interacted, the filament erupted. You can read more about the teams results in the article linked below. Also, check out this awesome video of the filament formation and eruption, again by SDO/AIA: Wang et al 2017 ApJ 839 128. doi:10.3847/1538-4357/aa6bf3

  20. Chordoid glioma with intraventricular dissemination: A case report with perfusion MR imaging features

    Ki, So Yeon; Kim, Seul Kee; Heo, Tae Wook; Baek, Byung Hyun; Kim, Hyung Seok; Yoon, Woong [Chonnam National University Medical School, Chonnam National University Hospital, Gwangju (Korea, Republic of)


    Chordoid glioma is a rare low grade tumor typically located in the third ventricle. Although a chordoid glioma can arise from ventricle with tumor cells having features of ependymal differentiation, intraventricular dissemination has not been reported. Here we report a case of a patient with third ventricular chordoid glioma and intraventricular dissemination in the lateral and fourth ventricles. We described the perfusion MR imaging features of our case different from a previous report.

  1. Utility of MRI features in differentiation of central renal cell carcinoma and renal pelvic urothelial carcinoma.

    Wehrli, Natasha E; Kim, Min Ju; Matza, Brent W; Melamed, Jonathan; Taneja, Samir S; Rosenkrantz, Andrew B


    The purpose of this article is to evaluate the utility of various morphologic and quantitative MRI features in differentiating central renal cell carcinoma (RCC) from renal pelvic urothelial carcinoma. Sixty patients (39 men and 21 women; mean [± SD] age, 65 ± 14 years; 48 with central RCC and 12 with renal pelvic urothelial carcinoma) who underwent MRI, including diffusion-weighted imaging (b values, 0, 400, and 800 s/mm(2)) and dynamic contrast-enhanced imaging, before histopathologic confirmation were included. Tumor T2 signal intensity and apparent diffusion coefficients (ADCs) were measured and normalized to muscle and CSF (hereafter referred to as normalized T2 signal and normalized ADC, respectively) and then were compared using receiver operating characteristic analysis. Also, two blinded radiologists independently assessed all tumors for various qualitative features, which were compared with the Fisher exact test and unpaired Student t test. Urothelial carcinoma exhibited significantly lower normalized ADC than did RCC (p = 0.008), but no significant difference was seen in ADC or normalized T2 signal intensity (p = 0.247-0.773). Normalized ADC had the highest area under the curve (0.757); normalized ADC below an optimal threshold of 0.451 was associated with sensitivity of 83% and specificity of 71% for diagnosing urothelial carcinoma. Features that were significantly more prevalent in urothelial carcinoma included global impression of urothelial carcinoma, location centered within the collecting system, collecting system defect, extension to the ureteropelvic junction, preserved renal shape, absence of cystic or necrotic areas, absence of hemorrhage, homogeneous enhancement, and hypovascularity (all p features ranged from 61.7% to 98.3%. In addition to various qualitative MRI parameters, normalized ADC has utility in differentiating central RCC from renal pelvic urothelial carcinoma. Such differentiation may assist decisions regarding possible biopsy

  2. Performance Analysis of Texture Image Classification Using Wavelet Feature

    Dolly Choudhary


    Full Text Available This paper compares the performance of various classifiers for multi class image classification. Where the features are extracted by the proposed algorithm in using Haar wavelet coefficient. The wavelet features are extracted from original texture images and corresponding complementary images. As it is really very difficult to decide which classifier would show better performance for multi class image classification. Hence, this work is an analytical study of performance of various classifiers for the single multiclass classification problem. In this work fifteen textures are taken for classification using Feed Forward Neural Network, Naïve Bays Classifier, K-nearest neighbor Classifier and Cascaded Neural Network.

  3. Automatic classification of hepatocellular carcinoma images based on nuclear and structural features

    Kiyuna, Tomoharu; Saito, Akira; Marugame, Atsushi; Yamashita, Yoshiko; Ogura, Maki; Cosatto, Eric; Abe, Tokiya; Hashiguchi, Akinori; Sakamoto, Michiie


    Diagnosis of hepatocellular carcinoma (HCC) on the basis of digital images is a challenging problem because, unlike gastrointestinal carcinoma, strong structural and morphological features are limited and sometimes absent from HCC images. In this study, we describe the classification of HCC images using statistical distributions of features obtained from image analysis of cell nuclei and hepatic trabeculae. Images of 130 hematoxylin-eosin (HE) stained histologic slides were captured at 20X by a slide scanner (Nanozoomer, Hamamatsu Photonics, Japan) and 1112 regions of interest (ROI) images were extracted for classification (551 negatives and 561 positives, including 113 well-differentiated positives). For a single nucleus, the following features were computed: area, perimeter, circularity, ellipticity, long and short axes of elliptic fit, contour complexity and gray level cooccurrence matrix (GLCM) texture features (angular second moment, contrast, homogeneity and entropy). In addition, distributions of nuclear density and hepatic trabecula thickness within an ROI were also extracted. To represent an ROI, statistical distributions (mean, standard deviation and percentiles) of these features were used. In total, 78 features were extracted for each ROI and a support vector machine (SVM) was trained to classify negative and positive ROIs. Experimental results using 5-fold cross validation show 90% sensitivity for an 87.8% specificity. The use of statistical distributions over a relatively large area makes the HCC classifier robust to occasional failures in the extraction of nuclear or hepatic trabecula features, thus providing stability to the system.


    A. Hema


    Full Text Available With the help of image mining techniques, an automatic pill identification system was investigated in this study for matching the images of the pills based on its several features like imprint, color, size and shape. Image mining is an inter-disciplinary task requiring expertise from various fields such as computer vision, image retrieval, image matching and pattern recognition. Image mining is the method in which the unusual patterns are detected so that both hidden and useful data images can only be stored in large database. It involves two different approaches for image matching. This research presents a drug identification, registration, detection and matching, Text, color and shape extraction of the image with image mining concept to identify the legal and illegal pills with more accuracy. Initially, the preprocessing process is carried out using novel interpolation algorithm. The main aim of this interpolation algorithm is to reduce the artifacts, blurring and jagged edges introduced during up-sampling. Then the registration process is proposed with two modules they are, feature extraction and corner detection. In feature extraction the noisy high frequency edges are discarded and relevant high frequency edges are selected. The corner detection approach detects the high frequency pixels in the intersection points. Through the overall performance gets improved. There is a need of segregate the dataset into groups based on the query image’s size, shape, color, text, etc. That process of segregating required information is called as feature extraction. The feature extraction is done using Geometrical Gradient feature transformation. Finally, color and shape feature extraction were performed using color histogram and geometrical gradient vector. Simulation results shows that the proposed techniques provide accurate retrieval results both in terms of time and accuracy when compared to conventional approaches.

  5. Review of Metaplastic Carcinoma of the Breast: Imaging Findings and Pathologic Features

    Rebecca Leddy


    Full Text Available Metaplastic carcinoma (MPC, an uncommon but often aggressive breast cancer, can be challenging to differentiate from other types of breast cancer and even benign lesions based on the imaging appearance. It has a variable pathology classification system. These types of tumors are generally rapidly growing palpable masses. MPCs on imaging can present with imaging features similar to invasive ductal carcinoma and probably even benign lesions. The purpose of this article is to review MPC of the breast including the pathology subtypes, imaging features, and imaging pathology correlations. By understanding the clinical picture, pathology, and overlap in imaging characteristics of MPC with invasive ductal carcinoma and probably benign lesions can assist in diagnosing these difficult malignancies.

  6. Feature Selection for Image Retrieval based on Genetic Algorithm

    Preeti Kushwaha


    Full Text Available This paper describes the development and implementation of feature selection for content based image retrieval. We are working on CBIR system with new efficient technique. In this system, we use multi feature extraction such as colour, texture and shape. The three techniques are used for feature extraction such as colour moment, gray level co- occurrence matrix and edge histogram descriptor. To reduce curse of dimensionality and find best optimal features from feature set using feature selection based on genetic algorithm. These features are divided into similar image classes using clustering for fast retrieval and improve the execution time. Clustering technique is done by k-means algorithm. The experimental result shows feature selection using GA reduces the time for retrieval and also increases the retrieval precision, thus it gives better and faster results as compared to normal image retrieval system. The result also shows precision and recall of proposed approach compared to previous approach for each image class. The CBIR system is more efficient and better performs using feature selection based on Genetic Algorithm.

  7. Ewing sarcoma versus osteomyelitis: differential diagnosis with magnetic resonance imaging

    Henninger, B.; Glodny, B.; Rudisch, A.; Trieb, T.; Loizides, A.; Judmaier, W.; Schocke, M.F. [Innsbruck Medical University, Department of Radiology, Innsbruck (Austria); Putzer, D. [Innsbruck Medical University, Department of Nuclear Medicine, Innsbruck (Austria)


    To find and evaluate characteristic magnetic resonance imaging (MRI) patterns for the differentiation between Ewing sarcoma and osteomyelitis. We identified 28 consecutive patients referred to our department for MRI (1.5 T) of an unclear bone lesion with clinical symptoms suggestive of Ewing sarcoma or osteomyelitis. MRI scans were re-evaluated by two experienced radiologists, typical MR imaging features were documented and a diagnostic decision between Ewing sarcoma and osteomyelitis was made. Statistical significance of the association between MRI features and the biopsy-based diagnosis was assessed using Fisher's exact test. The most clear-cut pattern for determining the correct diagnosis was the presence of a sharp and defined margin of the bone lesion, which was found in all patients with Ewing sarcoma, but in none of the patients with osteomyelitis (P < 0.0001). Contrast enhancing soft tissue was present in all cases with Ewing sarcoma and absent in 4 patients with osteomyelitis (P = 0.0103). Cortical destruction was found in all patients with Ewing sarcoma, 4 patients with osteomyelitis did not present any cortical reaction (P = 0.0103). Cystic or necrotic areas were identified in 13 patients with Ewing sarcoma and in 1 patient with osteomyelitis (P = 0.004). Interobserver reliability was very good (kappa = 1) in Ewing sarcoma and moderate (kappa = 0.6) in patients with osteomyelitis. A sharp and defined margin, optimally visualized on T1-weighted images in comparison to short tau inversion recovery (STIR) images, is the most significant feature of Ewing sarcoma in differentiating from osteomyelitis. (orig.)

  8. Ship Targets Discrimination Algorithm in SAR Images Based on Hu Moment Feature and Texture Feature

    Liu Lei


    Full Text Available To discriminate the ship targets in SAR images, this paper proposed the method based on combination of Hu moment feature and texture feature. Firstly,7 Hu moment features should be extracted, while gray level co-occurrence matrix is then used to extract the features of mean, variance, uniformity, energy, entropy, inertia moment, correlation and differences. Finally the k-neighbour classifier was used to analysis the 15 dimensional feature vectors. The experimental results show that the method of this paper has a good effect.

  9. Features Selection for Skin Micro-Image Symptomatic Recognition

    HU Yue-li; CAO Jia-lin; ZHAO Qian; FENG Xu


    Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom.This paper proposes a hybrid approach based on the support vector machine (SVM) technique and genetic algorithm (GA) to select an optimum feature subset from the feature group extracted from the skin micro-images. An adaptive GA is introduced for maintaining the convergence rate. With the proposed method, the average cross validation accuracy is increased from 88.25% using all features to 96.92 % using only selected features provided by a classifier for classification of 5 classes of skin symptoms. The experimental results are satisfactory.

  10. Features Selection for Skin Micro-Image Symptomatic Recognition

    HUYue-li; CAOJia-lin; ZHAOQian; FENGXu


    Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom.This paper proposes a hybrid approach based on the support vector machine (SVM) technique and genetic algorithm (GA) to select an optimum feature subset from the feature group extracted from the skin micro-images. An adaptive GA is introduced for maintaining the convergence rate. With the proposed method, the average cross validation accuracy is increased from 88.25% using all features to 96.92% using only selected features provided by a classifier for classification of 5 classes of skin symptoms. The experimental results are satisfactory.

  11. An adaptive multi-feature segmentation model for infrared image

    Zhang, Tingting; Han, Jin; Zhang, Yi; Bai, Lianfa


    Active contour models (ACM) have been extensively applied to image segmentation, conventional region-based active contour models only utilize global or local single feature information to minimize the energy functional to drive the contour evolution. Considering the limitations of original ACMs, an adaptive multi-feature segmentation model is proposed to handle infrared images with blurred boundaries and low contrast. In the proposed model, several essential local statistic features are introduced to construct a multi-feature signed pressure function (MFSPF). In addition, we draw upon the adaptive weight coefficient to modify the level set formulation, which is formed by integrating MFSPF with local statistic features and signed pressure function with global information. Experimental results demonstrate that the proposed method can make up for the inadequacy of the original method and get desirable results in segmenting infrared images.

  12. Segmentation of MR images using multiple-feature vectors

    Cole, Orlean I. B.; Daemi, Mohammad F.


    Segmentation is an important step in the analysis of MR images (MRI). Considerable progress has been made in this area, and numerous reports on 3D segmentation, volume measurement and visualization have been published in recent years. The main purpose of our study is to investigate the power and use of fractal techniques in extraction of features from MR images of the human brain. These features which are supplemented by other features are used for segmentation, and ultimately for the extraction of a known pathology, in our case multiple- sclerosis (MS) lesions. We are particularly interested in the progress of the lesions and occurrence of new lesions which in a typical case are scattered within the image and are sometimes difficult to identify visually. We propose a technique for multi-channel segmentation of MR images using multiple feature vectors. The channels are proton density, T1-weighted and T2-weighted images containing multiple-sclerosis (MS) lesions at various stages of development. We first represent each image as a set of feature vectors which are estimated using fractal techniques, and supplemented by micro-texture features and features from the gray-level co-occurrence matrix (GLCM). These feature vectors are then used in a feature selection algorithm to reduce the dimension of the feature space. The next stage is segmentation and clustering. The selected feature vectors now form the input to the segmentation and clustering routines and are used as the initial clustering parameters. For this purpose, we have used the classical K-means as the initial clustering method. The clustered image is then passed into a probabilistic classifier to further classify and validate each region, taking into account the spatial properties of the image. Initially, segmentation results were obtained using the fractal dimension features alone. Subsequently, a combination of the fractal dimension features and the supplementary features mentioned above were also obtained

  13. Image Retrieval Using a Combination of Keywords and Image Features

    Reddy, Vishwanath Reddy Keshi; Bandikolla, Praveen


    Information retrieval systems are playing an important role in our day to day life for getting the required information. Many text retrieval systems are available and are working successfully. Even though internet is full of other media like images, audio and video, retrieval systems for these media are rare and have not achieved success as that of text retrieval systems. Image retrieval systems are useful in many applications; there is a high demand for effective and efficient tool for image...

  14. Featured Image: The Birth of Spiral Arms

    Kohler, Susanna


    In this figure, the top panels show three spiral galaxies in the Virgo cluster, imaged with the Sloan Digital Sky Survey. The bottom panels provide a comparison with three morphologically similar galaxies generated insimulations. The simulations run by Marcin Semczuk, Ewa okas, and Andrs del Pino (Nicolaus Copernicus Astronomical Center, Poland) were designed to examine how the spiral arms of galaxies like the Milky Way may have formed. In particular, the group exploredthe possibility that so-called grand-design spiral arms are caused by tidal effects as a Milky-Way-like galaxy orbits a cluster of galaxies. The authors show that the gravitational potential of the cluster can trigger the formation of two spiral arms each time the galaxy passes through the pericenter of its orbit around the cluster. Check out the original paper below for more information!CitationMarcin Semczuk et al 2017 ApJ 834 7. doi:10.3847/1538-4357/834/1/7

  15. Image processing tool for automatic feature recognition and quantification

    Chen, Xing; Stoddard, Ryan J.


    A system for defining structures within an image is described. The system includes reading of an input file, preprocessing the input file while preserving metadata such as scale information and then detecting features of the input file. In one version the detection first uses an edge detector followed by identification of features using a Hough transform. The output of the process is identified elements within the image.

  16. Image Mosaic Method Based on SIFT Features of Line Segment

    Jun Zhu


    Full Text Available This paper proposes a novel image mosaic method based on SIFT (Scale Invariant Feature Transform feature of line segment, aiming to resolve incident scaling, rotation, changes in lighting condition, and so on between two images in the panoramic image mosaic process. This method firstly uses Harris corner detection operator to detect key points. Secondly, it constructs directed line segments, describes them with SIFT feature, and matches those directed segments to acquire rough point matching. Finally, Ransac method is used to eliminate wrong pairs in order to accomplish image mosaic. The results from experiment based on four pairs of images show that our method has strong robustness for resolution, lighting, rotation, and scaling.

  17. Breast image feature learning with adaptive deconvolutional networks

    Jamieson, Andrew R.; Drukker, Karen; Giger, Maryellen L.


    Feature extraction is a critical component of medical image analysis. Many computer-aided diagnosis approaches employ hand-designed, heuristic lesion extracted features. An alternative approach is to learn features directly from images. In this preliminary study, we explored the use of Adaptive Deconvolutional Networks (ADN) for learning high-level features in diagnostic breast mass lesion images with potential application to computer-aided diagnosis (CADx) and content-based image retrieval (CBIR). ADNs (Zeiler, et. al., 2011), are recently-proposed unsupervised, generative hierarchical models that decompose images via convolution sparse coding and max pooling. We trained the ADNs to learn multiple layers of representation for two breast image data sets on two different modalities (739 full field digital mammography (FFDM) and 2393 ultrasound images). Feature map calculations were accelerated by use of GPUs. Following Zeiler et. al., we applied the Spatial Pyramid Matching (SPM) kernel (Lazebnik, et. al., 2006) on the inferred feature maps and combined this with a linear support vector machine (SVM) classifier for the task of binary classification between cancer and non-cancer breast mass lesions. Non-linear, local structure preserving dimension reduction, Elastic Embedding (Carreira-Perpiñán, 2010), was then used to visualize the SPM kernel output in 2D and qualitatively inspect image relationships learned. Performance was found to be competitive with current CADx schemes that use human-designed features, e.g., achieving a 0.632+ bootstrap AUC (by case) of 0.83 [0.78, 0.89] for an ultrasound image set (1125 cases).

  18. The analysis of image feature robustness using cometcloud

    Xin Qi


    Full Text Available The robustness of image features is a very important consideration in quantitative image analysis. The objective of this paper is to investigate the robustness of a range of image texture features using hematoxylin stained breast tissue microarray slides which are assessed while simulating different imaging challenges including out of focus, changes in magnification and variations in illumination, noise, compression, distortion, and rotation. We employed five texture analysis methods and tested them while introducing all of the challenges listed above. The texture features that were evaluated include co-occurrence matrix, center-symmetric auto-correlation, texture feature coding method, local binary pattern, and texton. Due to the independence of each transformation and texture descriptor, a network structured combination was proposed and deployed on the Rutgers private cloud. The experiments utilized 20 randomly selected tissue microarray cores. All the combinations of the image transformations and deformations are calculated, and the whole feature extraction procedure was completed in 70 minutes using a cloud equipped with 20 nodes. Center-symmetric auto-correlation outperforms all the other four texture descriptors but also requires the longest computational time. It is roughly 10 times slower than local binary pattern and texton. From a speed perspective, both the local binary pattern and texton features provided excellent performance for classification and content-based image retrieval.

  19. Differential Electrochemical Conductance Imaging at the Nanoscale.

    López-Martínez, Montserrat; Artés, Juan Manuel; Sarasso, Veronica; Carminati, Marco; Díez-Pérez, Ismael; Sanz, Fausto; Gorostiza, Pau


    Electron transfer in proteins is essential in crucial biological processes. Although the fundamental aspects of biological electron transfer are well characterized, currently there are no experimental tools to determine the atomic-scale electronic pathways in redox proteins, and thus to fully understand their outstanding efficiency and environmental adaptability. This knowledge is also required to design and optimize biomolecular electronic devices. In order to measure the local conductance of an electrode surface immersed in an electrolyte, this study builds upon the current-potential spectroscopic capacity of electrochemical scanning tunneling microscopy, by adding an alternating current modulation technique. With this setup, spatially resolved, differential electrochemical conductance images under bipotentiostatic control are recorded. Differential electrochemical conductance imaging allows visualizing the reversible oxidation of an iron electrode in borate buffer and individual azurin proteins immobilized on atomically flat gold surfaces. In particular, this method reveals submolecular regions with high conductance within the protein. The direct observation of nanoscale conduction pathways in redox proteins and complexes enables important advances in biochemistry and bionanotechnology. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

  20. [Feature extraction for breast cancer data based on geometric algebra theory and feature selection using differential evolution].

    Li, Jing; Hong, Wenxue


    The feature extraction and feature selection are the important issues in pattern recognition. Based on the geometric algebra representation of vector, a new feature extraction method using blade coefficient of geometric algebra was proposed in this study. At the same time, an improved differential evolution (DE) feature selection method was proposed to solve the elevated high dimension issue. The simple linear discriminant analysis was used as the classifier. The result of the 10-fold cross-validation (10 CV) classification of public breast cancer biomedical dataset was more than 96% and proved superior to that of the original features and traditional feature extraction method.

  1. An efficient approach for differentiating Alzheimer's disease from normal elderly based on multicenter MRI using gray-level invariant features.

    Muwei Li

    Full Text Available Machine learning techniques, along with imaging markers extracted from structural magnetic resonance images, have been shown to increase the accuracy to differentiate patients with Alzheimer's disease (AD from normal elderly controls. Several forms of anatomical features, such as cortical volume, shape, and thickness, have demonstrated discriminative capability. These approaches rely on accurate non-linear image transformation, which could invite several nuisance factors, such as dependency on transformation parameters and the degree of anatomical abnormality, and an unpredictable influence of residual registration errors. In this study, we tested a simple method to extract disease-related anatomical features, which is suitable for initial stratification of the heterogeneous patient populations often encountered in clinical data. The method employed gray-level invariant features, which were extracted from linearly transformed images, to characterize AD-specific anatomical features. The intensity information from a disease-specific spatial masking, which was linearly registered to each patient, was used to capture the anatomical features. We implemented a two-step feature selection for anatomic recognition. First, a statistic-based feature selection was implemented to extract AD-related anatomical features while excluding non-significant features. Then, seven knowledge-based ROIs were used to capture the local discriminative powers of selected voxels within areas that were sensitive to AD or mild cognitive impairment (MCI. The discriminative capability of the proposed feature was measured by its performance in differentiating AD or MCI from normal elderly controls (NC using a support vector machine. The statistic-based feature selection, together with the knowledge-based masks, provided a promising solution for capturing anatomical features of the brain efficiently. For the analysis of clinical populations, which are inherently heterogeneous

  2. Featured Image: Mapping Jupiter with Hubble

    Kohler, Susanna


    Zonal wind profile for Jupiter, describing the speed and direction of its winds at each latitude. [Simon et al. 2015]This global map of Jupiters surface (click for the full view!) was generated by the Hubble Outer Planet Atmospheres Legacy (OPAL) program, which aims to createnew yearly global maps for each of the outer planets. Presented in a study led by Amy Simon (NASA Goddard Space Flight Center), the map above is the first generated for Jupiter in the first year of the OPAL campaign. It provides a detailed look at Jupiters atmospheric structure including the Great Red Spot and allowed the authors to measure the speed and direction of the wind across Jupiters latitudes, constructing an updated zonal wind profile for Jupiter.In contrast to this study, the Juno mission (which will be captured into Jupiters orbit today after a 5-year journey to Jupiter!) will be focusing more on the features below Jupiters surface, studying its deep atmosphere and winds. Some of Junos primary goals are to learn about Jupiters composition, gravitational field, magnetic field, and polar magnetosphere. You can follow along with the NASATV livestream as Juno arrives at Jupiter tonight; orbit insertion coverage starts at 10:30 EDT.CitationAmy A. Simon et al 2015 ApJ 812 55. doi:10.1088/0004-637X/812/1/55

  3. Automated blood vessel extraction using local features on retinal images

    Hatanaka, Yuji; Samo, Kazuki; Tajima, Mikiya; Ogohara, Kazunori; Muramatsu, Chisako; Okumura, Susumu; Fujita, Hiroshi


    An automated blood vessel extraction using high-order local autocorrelation (HLAC) on retinal images is presented. Although many blood vessel extraction methods based on contrast have been proposed, a technique based on the relation of neighbor pixels has not been published. HLAC features are shift-invariant; therefore, we applied HLAC features to retinal images. However, HLAC features are weak to turned image, thus a method was improved by the addition of HLAC features to a polar transformed image. The blood vessels were classified using an artificial neural network (ANN) with HLAC features using 105 mask patterns as input. To improve performance, the second ANN (ANN2) was constructed by using the green component of the color retinal image and the four output values of ANN, Gabor filter, double-ring filter and black-top-hat transformation. The retinal images used in this study were obtained from the "Digital Retinal Images for Vessel Extraction" (DRIVE) database. The ANN using HLAC output apparent white values in the blood vessel regions and could also extract blood vessels with low contrast. The outputs were evaluated using the area under the curve (AUC) based on receiver operating characteristics (ROC) analysis. The AUC of ANN2 was 0.960 as a result of our study. The result can be used for the quantitative analysis of the blood vessels.

  4. Convolutional neural network features based change detection in satellite images

    Mohammed El Amin, Arabi; Liu, Qingjie; Wang, Yunhong


    With the popular use of high resolution remote sensing (HRRS) satellite images, a huge research efforts have been placed on change detection (CD) problem. An effective feature selection method can significantly boost the final result. While hand-designed features have proven difficulties to design features that effectively capture high and mid-level representations, the recent developments in machine learning (Deep Learning) omit this problem by learning hierarchical representation in an unsupervised manner directly from data without human intervention. In this letter, we propose approaching the change detection problem from a feature learning perspective. A novel deep Convolutional Neural Networks (CNN) features based HR satellite images change detection method is proposed. The main guideline is to produce a change detection map directly from two images using a pretrained CNN. This method can omit the limited performance of hand-crafted features. Firstly, CNN features are extracted through different convolutional layers. Then, a concatenation step is evaluated after an normalization step, resulting in a unique higher dimensional feature map. Finally, a change map was computed using pixel-wise Euclidean distance. Our method has been validated on real bitemporal HRRS satellite images according to qualitative and quantitative analyses. The results obtained confirm the interest of the proposed method.

  5. Analysis of mammogram images based on texture features of curvelet sub-bands

    Gardezi, Syed Jamal Safdar; Faye, Ibrahima; Eltoukhy, Mohamed Meselhy


    Image texture analysis plays an important role in object detection and recognition in image processing. The texture analysis can be used for early detection of breast cancer by classifying the mammogram images into normal and abnormal classes. This study investigates breast cancer detection using texture features obtained from the grey level cooccurrence matrices (GLCM) of curvelet sub-band levels combined with texture feature obtained from the image itself. The GLCM were constructed for each sub-band of three curvelet decomposition levels. The obtained feature vector presented to the classifier to differentiate between normal and abnormal tissues. The proposed method is applied over 305 region of interest (ROI) cropped from MIAS dataset. The simple logistic classifier achieved 86.66% classification accuracy rate with sensitivity 76.53% and specificity 91.3%.

  6. Featured Image: A Looping Stellar Stream

    Kohler, Susanna


    This negative image of NGC 5907 (originally published inMartinez-Delgadoet al. 2008; click for the full view!) reveals the faint stellar stream that encircles the galaxy, forming loops around it a fossil of a recent merger. Mergers between galaxies come in several different flavors: major mergers, in which the merging galaxies are within a 1:5 ratio in stellar mass; satellite cannibalism, in which a large galaxy destroys a small satellite less than a 50th of its size; and the in-between case of minor mergers, in which the merging galaxieshave stellar mass ratios between 1:5 and 1:50. These minor mergers are thought to be relatively common, and they can have a significant effect on the dynamics and structure of the primary galaxy. A team of scientists led by Seppo Laine (Spitzer Science Center Caltech) has recently analyzed the metallicity and age of the stellar population in the stream around NGC 5907. By fitting these observations with a stellar population synthesis model, they conclude that this stream is an example of a massive minor merger, with a stellar mass ratio of at least 1:8. For more information, check out the paper below!CitationSeppo Laine et al 2016 AJ 152 72. doi:10.3847/0004-6256/152/3/72

  7. Featured Image: Fireball After a Temporary Capture?

    Kohler, Susanna


    This image of a fireball was captured in the Czech Republic by cameras at a digital autonomous observatory in the village of Kunak. This observatory is part of a network of stations known as the European Fireball Network, and this particular meteoroid detection, labeled EN130114, is notable because it has the lowest initial velocity of any natural object ever observed by the network. Led by David Clark (University of Western Ontario), the authors of a recent study speculate that before this meteoroid impacted Earth, it may have been a Temporarily Captured Orbiter (TCO). TCOs are near-Earth objects that make a few orbits of Earth before returning to heliocentric orbits. Only one has ever been observed to date, and though they are thought to make up 0.1% of all meteoroids, EN130114 is the first event ever detected that exhibits conclusive behavior of a TCO. For more information on EN130114 and why TCOs are important to study, check out the paper below!CitationDavid L. Clark et al 2016 AJ 151 135. doi:10.3847/0004-6256/151/6/135

  8. Featured Image: The Cosmic Velocity Web

    Kohler, Susanna


    You may have heard of the cosmic web, a network of filaments, clusters and voids that describes the three-dimensional distribution of matter in our universe. But have you ever considered the idea of a cosmic velocity web? In a new study led by Daniel Pomarde (IRFU CEA-Saclay, France), a team of scientists has built a detailed 3D view of the flows in our universe, showing in particular motions along filaments and in collapsing knots. In the image above (click for the full view), surfaces of knots (red) are embedded within surfaces of filaments (grey). The rainbow lines show the flow motion, revealing acceleration (redder tones) toward knots and retardation (bluer tones) beyond them. You can learn more about Pomarde and collaborators work and see their unusual and intriguing visualizationsin the video they produced, below. Check out the original paper for more information.CitationDaniel Pomarde et al 2017 ApJ 845 55. doi:10.3847/1538-4357/aa7f78

  9. Featured Image: Experimental Simulation of Melting Meteoroids

    Kohler, Susanna


    Ever wonder what experimental astronomy looks like? Some days, it looks like this piece of rock in a wind tunnel (click for a betterlook!). In this photo, a piece of agrillite (a terrestrial rock) is exposed to conditions in a plasma wind tunnel as a team of scientists led by Stefan Loehle (Stuttgart University) simulate what happens to a meteoroid as it hurtles through Earths atmosphere. With these experiments, the scientists hope to better understand meteoroid ablation the process by which meteoroids are heated, melt, and evaporateas they pass through our atmosphere so that we can learn more from the meteorite fragments that make it to the ground. In the scientists experiment, the rock samples were exposed to plasma flow until they disintegrated, and this process was simultaneously studied via photography, video, high-speed imaging, thermography, and Echelle emission spectroscopy. To find out what the team learned from these experiments, you can check out the original article below.CitationStefan Loehle et al 2017 ApJ 837 112. doi:10.3847/1538-4357/aa5cb5

  10. Robust Tomato Recognition for Robotic Harvesting Using Feature Images Fusion

    Yuanshen Zhao


    Full Text Available Automatic recognition of mature fruits in a complex agricultural environment is still a challenge for an autonomous harvesting robot due to various disturbances existing in the background of the image. The bottleneck to robust fruit recognition is reducing influence from two main disturbances: illumination and overlapping. In order to recognize the tomato in the tree canopy using a low-cost camera, a robust tomato recognition algorithm based on multiple feature images and image fusion was studied in this paper. Firstly, two novel feature images, the  a*-component image and the I-component image, were extracted from the L*a*b* color space and luminance, in-phase, quadrature-phase (YIQ color space, respectively. Secondly, wavelet transformation was adopted to fuse the two feature images at the pixel level, which combined the feature information of the two source images. Thirdly, in order to segment the target tomato from the background, an adaptive threshold algorithm was used to get the optimal threshold. The final segmentation result was processed by morphology operation to reduce a small amount of noise. In the detection tests, 93% target tomatoes were recognized out of 200 overall samples. It indicates that the proposed tomato recognition method is available for robotic tomato harvesting in the uncontrolled environment with low cost.

  11. Multiresolution image fusion scheme based on fuzzy region feature

    LIU Gang; JING Zhong-liang; SUN Shao-yuan


    This paper proposes a novel region based image fusion scheme based on multiresolution analysis. The low frequency band of the image multiresolution representation is segmented into important regions, sub-important regions and background regions. Each feature of the regions is used to determine the region's degree of membership in the multiresolution representation,and then to achieve multiresolution representation of the fusion result. The final image fusion result can be obtained by using the inverse multiresolution transform. Experiments showed that the proposed image fusion method can have better performance than existing image fusion methods.

  12. Disorders of the pediatric pancreas: imaging features

    Nijs, Els [University Hospital Gasthuisberg, Department of Radiology, Leuven (Belgium); Callahan, Michael J.; Taylor, George A. [Boston Children' s Hospital, Department of Radiology, Boston, MA (United States)


    The purpose of this manuscript is to provide an overview of the normal development of the pancreas as well as pancreatic pathology in children. Diagnostic imaging plays a major role in the evaluation of the pancreas in infants and children. Familiarity with the range of normal appearance and the diseases that commonly affect this gland is important for the accurate and timely diagnosis of pancreatic disorders in the pediatric population. Normal embryology is discussed, as are the most common congenital anomalies that occur as a result of aberrant development during embryology. These include pancreas divisum, annular pancreas, agenesis of the dorsal pancreatic anlagen and ectopic pancreatic tissue. Syndromes that can manifest pancreatic pathology include: Beckwith Wiedemann syndrome, von Hippel-Lindau disease and autosomal dominant polycystic kidney disease. Children and adults with cystic fibrosis and Shwachman-Diamond syndrome frequently present with pancreatic insufficiency. Trauma is the most common cause of pancreatitis in children. In younger children, unexplained pancreatic injury must always alert the radiologist to potential child abuse. Pancreatic pseudocysts are a complication of trauma, but can also be seen in the setting of acute or chronic pancreatitis from other causes. Primary pancreatic neoplasms are rare in children and are divided into exocrine tumors such as pancreatoblastoma and adenocarcinoma and into endocrine or islet cell tumors. Islet cell tumors are classified as functioning (insulinoma, gastrinoma, VIPoma and glucagonoma) and nonfunctioning tumors. Solid-cystic papillary tumor is probably the most common pancreatic tumor in Asian children. Although quite rare, secondary tumors of the pancreas can be associated with certain primary malignancies. (orig.)

  13. Image Processing Techniques and Feature Recognition in Solar Physics

    Aschwanden, Markus J.


    This review presents a comprehensive and systematic overview of image-processing techniques that are used in automated feature-detection algorithms applied to solar data: i) image pre-processing procedures, ii) automated detection of spatial features, iii) automated detection and tracking of temporal features (events), and iv) post-processing tasks, such as visualization of solar imagery, cataloguing, statistics, theoretical modeling, prediction, and forecasting. For each aspect the most recent developments and science results are highlighted. We conclude with an outlook on future trends.

  14. Integration of Image-Derived and Pos-Derived Features for Image Blur Detection

    Teo, Tee-Ann; Zhan, Kai-Zhi


    The image quality plays an important role for Unmanned Aerial Vehicle (UAV)'s applications. The small fixed wings UAV is suffering from the image blur due to the crosswind and the turbulence. Position and Orientation System (POS), which provides the position and orientation information, is installed onto an UAV to enable acquisition of UAV trajectory. It can be used to calculate the positional and angular velocities when the camera shutter is open. This study proposes a POS-assisted method to detect the blur image. The major steps include feature extraction, blur image detection and verification. In feature extraction, this study extracts different features from images and POS. The image-derived features include mean and standard deviation of image gradient. For POS-derived features, we modify the traditional degree-of-linear-blur (blinear) method to degree-of-motion-blur (bmotion) based on the collinear condition equations and POS parameters. Besides, POS parameters such as positional and angular velocities are also adopted as POS-derived features. In blur detection, this study uses Support Vector Machines (SVM) classifier and extracted features (i.e. image information, POS data, blinear and bmotion) to separate blur and sharp UAV images. The experiment utilizes SenseFly eBee UAV system. The number of image is 129. In blur image detection, we use the proposed degree-of-motion-blur and other image features to classify the blur image and sharp images. The classification result shows that the overall accuracy using image features is only 56%. The integration of image-derived and POS-derived features have improved the overall accuracy from 56% to 76% in blur detection. Besides, this study indicates that the performance of the proposed degree-of-motion-blur is better than the traditional degree-of-linear-blur.

  15. Face recognition with multi-resolution spectral feature images.

    Zhan-Li Sun

    Full Text Available The one-sample-per-person problem has become an active research topic for face recognition in recent years because of its challenges and significance for real-world applications. However, achieving relatively higher recognition accuracy is still a difficult problem due to, usually, too few training samples being available and variations of illumination and expression. To alleviate the negative effects caused by these unfavorable factors, in this paper we propose a more accurate spectral feature image-based 2DLDA (two-dimensional linear discriminant analysis ensemble algorithm for face recognition, with one sample image per person. In our algorithm, multi-resolution spectral feature images are constructed to represent the face images; this can greatly enlarge the training set. The proposed method is inspired by our finding that, among these spectral feature images, features extracted from some orientations and scales using 2DLDA are not sensitive to variations of illumination and expression. In order to maintain the positive characteristics of these filters and to make correct category assignments, the strategy of classifier committee learning (CCL is designed to combine the results obtained from different spectral feature images. Using the above strategies, the negative effects caused by those unfavorable factors can be alleviated efficiently in face recognition. Experimental results on the standard databases demonstrate the feasibility and efficiency of the proposed method.

  16. Face recognition with multi-resolution spectral feature images.

    Sun, Zhan-Li; Lam, Kin-Man; Dong, Zhao-Yang; Wang, Han; Gao, Qing-Wei; Zheng, Chun-Hou


    The one-sample-per-person problem has become an active research topic for face recognition in recent years because of its challenges and significance for real-world applications. However, achieving relatively higher recognition accuracy is still a difficult problem due to, usually, too few training samples being available and variations of illumination and expression. To alleviate the negative effects caused by these unfavorable factors, in this paper we propose a more accurate spectral feature image-based 2DLDA (two-dimensional linear discriminant analysis) ensemble algorithm for face recognition, with one sample image per person. In our algorithm, multi-resolution spectral feature images are constructed to represent the face images; this can greatly enlarge the training set. The proposed method is inspired by our finding that, among these spectral feature images, features extracted from some orientations and scales using 2DLDA are not sensitive to variations of illumination and expression. In order to maintain the positive characteristics of these filters and to make correct category assignments, the strategy of classifier committee learning (CCL) is designed to combine the results obtained from different spectral feature images. Using the above strategies, the negative effects caused by those unfavorable factors can be alleviated efficiently in face recognition. Experimental results on the standard databases demonstrate the feasibility and efficiency of the proposed method.

  17. An image segmentation based method for iris feature extraction

    XU Guang-zhu; ZHANG Zai-feng; MA Yi-de


    In this article, the local anomalistic blocks such ascrypts, furrows, and so on in the iris are initially used directly asiris features. A novel image segmentation method based onintersecting cortical model (ICM) neural network was introducedto segment these anomalistic blocks. First, the normalized irisimage was put into ICM neural network after enhancement.Second, the iris features were segmented out perfectly and wereoutput in binary image type by the ICM neural network. Finally,the fourth output pulse image produced by ICM neural networkwas chosen as the iris code for the convenience of real timeprocessing. To estimate the performance of the presentedmethod, an iris recognition platform was produced and theHamming Distance between two iris codes was computed tomeasure the dissimilarity between them. The experimentalresults in CASIA vl.0 and Bath iris image databases show thatthe proposed iris feature extraction algorithm has promisingpotential in iris recognition.

  18. Feature preserving compression of high resolution SAR images

    Yang, Zhigao; Hu, Fuxiang; Sun, Tao; Qin, Qianqing


    Compression techniques are required to transmit the large amounts of high-resolution synthetic aperture radar (SAR) image data over the available channels. Common Image compression methods may lose detail and weak information in original images, especially at smoothness areas and edges with low contrast. This is known as "smoothing effect". It becomes difficult to extract and recognize some useful image features such as points and lines. We propose a new SAR image compression algorithm that can reduce the "smoothing effect" based on adaptive wavelet packet transform and feature-preserving rate allocation. For the reason that images should be modeled as non-stationary information resources, a SAR image is partitioned to overlapped blocks. Each overlapped block is then transformed by adaptive wavelet packet according to statistical features of different blocks. In quantifying and entropy coding of wavelet coefficients, we integrate feature-preserving technique. Experiments show that quality of our algorithm up to 16:1 compression ratio is improved significantly, and more weak information is reserved.

  19. Histopathological Image Classification Using Discriminative Feature-Oriented Dictionary Learning.

    Vu, Tiep Huu; Mousavi, Hojjat Seyed; Monga, Vishal; Rao, Ganesh; Rao, U K Arvind


    In histopathological image analysis, feature extraction for classification is a challenging task due to the diversity of histology features suitable for each problem as well as presence of rich geometrical structures. In this paper, we propose an automatic feature discovery framework via learning class-specific dictionaries and present a low-complexity method for classification and disease grading in histopathology. Essentially, our Discriminative Feature-oriented Dictionary Learning (DFDL) method learns class-specific dictionaries such that under a sparsity constraint, the learned dictionaries allow representing a new image sample parsimoniously via the dictionary corresponding to the class identity of the sample. At the same time, the dictionary is designed to be poorly capable of representing samples from other classes. Experiments on three challenging real-world image databases: 1) histopathological images of intraductal breast lesions, 2) mammalian kidney, lung and spleen images provided by the Animal Diagnostics Lab (ADL) at Pennsylvania State University, and 3) brain tumor images from The Cancer Genome Atlas (TCGA) database, reveal the merits of our proposal over state-of-the-art alternatives. Moreover, we demonstrate that DFDL exhibits a more graceful decay in classification accuracy against the number of training images which is highly desirable in practice where generous training is often not available.

  20. Perceptual image hashing via feature points: performance evaluation and tradeoffs.

    Monga, Vishal; Evans, Brian L


    We propose an image hashing paradigm using visually significant feature points. The feature points should be largely invariant under perceptually insignificant distortions. To satisfy this, we propose an iterative feature detector to extract significant geometry preserving feature points. We apply probabilistic quantization on the derived features to introduce randomness, which, in turn, reduces vulnerability to adversarial attacks. The proposed hash algorithm withstands standard benchmark (e.g., Stirmark) attacks, including compression, geometric distortions of scaling and small-angle rotation, and common signal-processing operations. Content changing (malicious) manipulations of image data are also accurately detected. Detailed statistical analysis in the form of receiver operating characteristic (ROC) curves is presented and reveals the success of the proposed scheme in achieving perceptual robustness while avoiding misclassification.

  1. Image segmentation using an improved differential algorithm

    Gao, Hao; Shi, Yujiao; Wu, Dongmei


    Among all the existing segmentation techniques, the thresholding technique is one of the most popular due to its simplicity, robustness, and accuracy (e.g. the maximum entropy method, Otsu's method, and K-means clustering). However, the computation time of these algorithms grows exponentially with the number of thresholds due to their exhaustive searching strategy. As a population-based optimization algorithm, differential algorithm (DE) uses a population of potential solutions and decision-making processes. It has shown considerable success in solving complex optimization problems within a reasonable time limit. Thus, applying this method into segmentation algorithm should be a good choice during to its fast computational ability. In this paper, we first propose a new differential algorithm with a balance strategy, which seeks a balance between the exploration of new regions and the exploitation of the already sampled regions. Then, we apply the new DE into the traditional Otsu's method to shorten the computation time. Experimental results of the new algorithm on a variety of images show that, compared with the EA-based thresholding methods, the proposed DE algorithm gets more effective and efficient results. It also shortens the computation time of the traditional Otsu method.

  2. Image Recommendation Algorithm Using Feature-Based Collaborative Filtering

    Kim, Deok-Hwan

    As the multimedia contents market continues its rapid expansion, the amount of image contents used in mobile phone services, digital libraries, and catalog service is increasing remarkably. In spite of this rapid growth, users experience high levels of frustration when searching for the desired image. Even though new images are profitable to the service providers, traditional collaborative filtering methods cannot recommend them. To solve this problem, in this paper, we propose feature-based collaborative filtering (FBCF) method to reflect the user's most recent preference by representing his purchase sequence in the visual feature space. The proposed approach represents the images that have been purchased in the past as the feature clusters in the multi-dimensional feature space and then selects neighbors by using an inter-cluster distance function between their feature clusters. Various experiments using real image data demonstrate that the proposed approach provides a higher quality recommendation and better performance than do typical collaborative filtering and content-based filtering techniques.

  3. Hybrid edge and feature-based single-image superresolution

    Islam, Mohammad Moinul; Islam, Mohammed Nazrul; Asari, Vijayan K.; Karim, Mohammad A.


    A neighborhood-dependent component feature learning method for regression analysis in single-image superresolution is presented. Given a low-resolution input, the method uses a directional Fourier phase feature component to adaptively learn the regression kernel based on local covariance to estimate the high-resolution image. The unique feature of the proposed method is that it uses image features to learn about the local covariance from geometric similarity between the low-resolution image and its high-resolution counterpart. For each patch in the neighborhood, we estimate four directional variances to adapt the interpolated pixels. This gives us edge information and Fourier phase gives features, which are combined to interpolate using kernel regression. In order to compare quantitatively with other state-of-the-art techniques, root-mean-square error and measure mean-square similarity are computed for the example images, and experimental results show that the proposed algorithm outperforms similar techniques available in the literature, especially at higher resolution scales.

  4. MR imaging evaluation of perianal fistulas: spectrum of imaging features.

    de Miguel Criado, Jaime; del Salto, Laura García; Rivas, Patricia Fraga; del Hoyo, Luis Felipe Aguilera; Velasco, Leticia Gutiérrez; de las Vacas, M Isabel Díez Pérez; Marco Sanz, Ana G; Paradela, Marcos Manzano; Moreno, Eduardo Fraile


    Perianal fistulization is an inflammatory condition that affects the region around the anal canal, causing significant morbidity and often requiring repeated surgical treatments due to its high tendency to recur. To adopt the best surgical strategy and avoid recurrences, it is necessary to obtain precise radiologic information about the location of the fistulous track and the affected pelvic structures. Until recently, imaging techniques played a limited role in evaluation of perianal fistulas. However, magnetic resonance (MR) imaging now provides more precise information on the anatomy of the anal canal, the anal sphincter complex, and the relationships of the fistula to the pelvic floor structures and the plane of the levator ani muscle. MR imaging allows precise definition of the fistulous track and identification of secondary fistulas or abscesses. It provides accurate information for appropriate surgical treatment, decreasing the incidence of recurrence and allowing side effects such as fecal incontinence to be avoided. Radiologists should be familiar with the anatomic and pathologic findings of perianal fistulas and classify them using the St James's University Hospital MR imaging-based grading system.

  5. Simultenious binary hash and features learning for image retrieval

    Frantc, V. A.; Makov, S. V.; Voronin, V. V.; Marchuk, V. I.; Semenishchev, E. A.; Egiazarian, K. O.; Agaian, S.


    Content-based image retrieval systems have plenty of applications in modern world. The most important one is the image search by query image or by semantic description. Approaches to this problem are employed in personal photo-collection management systems, web-scale image search engines, medical systems, etc. Automatic analysis of large unlabeled image datasets is virtually impossible without satisfactory image-retrieval technique. It's the main reason why this kind of automatic image processing has attracted so much attention during recent years. Despite rather huge progress in the field, semantically meaningful image retrieval still remains a challenging task. The main issue here is the demand to provide reliable results in short amount of time. This paper addresses the problem by novel technique for simultaneous learning of global image features and binary hash codes. Our approach provide mapping of pixel-based image representation to hash-value space simultaneously trying to save as much of semantic image content as possible. We use deep learning methodology to generate image description with properties of similarity preservation and statistical independence. The main advantage of our approach in contrast to existing is ability to fine-tune retrieval procedure for very specific application which allow us to provide better results in comparison to general techniques. Presented in the paper framework for data- dependent image hashing is based on use two different kinds of neural networks: convolutional neural networks for image description and autoencoder for feature to hash space mapping. Experimental results confirmed that our approach has shown promising results in compare to other state-of-the-art methods.

  6. Modeling neuron selectivity over simple midlevel features for image classification.

    Shu Kong; Zhuolin Jiang; Qiang Yang


    We now know that good mid-level features can greatly enhance the performance of image classification, but how to efficiently learn the image features is still an open question. In this paper, we present an efficient unsupervised midlevel feature learning approach (MidFea), which only involves simple operations, such as k-means clustering, convolution, pooling, vector quantization, and random projection. We show this simple feature can also achieve good performance in traditional classification task. To further boost the performance, we model the neuron selectivity (NS) principle by building an additional layer over the midlevel features prior to the classifier. The NS-layer learns category-specific neurons in a supervised manner with both bottom-up inference and top-down analysis, and thus supports fast inference for a query image. Through extensive experiments, we demonstrate that this higher level NS-layer notably improves the classification accuracy with our simple MidFea, achieving comparable performances for face recognition, gender classification, age estimation, and object categorization. In particular, our approach runs faster in inference by an order of magnitude than sparse coding-based feature learning methods. As a conclusion, we argue that not only do carefully learned features (MidFea) bring improved performance, but also a sophisticated mechanism (NS-layer) at higher level boosts the performance further.

  7. Optimized Image Steganalysis through Feature Selection using MBEGA

    Geetha, S


    Feature based steganalysis, an emerging branch in information forensics, aims at identifying the presence of a covert communication by employing the statistical features of the cover and stego image as clues/evidences. Due to the large volumes of security audit data as well as complex and dynamic properties of steganogram behaviours, optimizing the performance of steganalysers becomes an important open problem. This paper is focussed at fine tuning the performance of six promising steganalysers in this field, through feature selection. We propose to employ Markov Blanket-Embedded Genetic Algorithm (MBEGA) for stego sensitive feature selection process. In particular, the embedded Markov blanket based memetic operators add or delete features (or genes) from a genetic algorithm (GA) solution so as to quickly improve the solution and fine-tune the search. Empirical results suggest that MBEGA is effective and efficient in eliminating irrelevant and redundant features based on both Markov blanket and predictive pow...

  8. MRI and PET image fusion using fuzzy logic and image local features.

    Javed, Umer; Riaz, Muhammad Mohsin; Ghafoor, Abdul; Ali, Syed Sohaib; Cheema, Tanveer Ahmed


    An image fusion technique for magnetic resonance imaging (MRI) and positron emission tomography (PET) using local features and fuzzy logic is presented. The aim of proposed technique is to maximally combine useful information present in MRI and PET images. Image local features are extracted and combined with fuzzy logic to compute weights for each pixel. Simulation results show that the proposed scheme produces significantly better results compared to state-of-art schemes.

  9. Optimization of wavelet decomposition for image compression and feature preservation.

    Lo, Shih-Chung B; Li, Huai; Freedman, Matthew T


    A neural-network-based framework has been developed to search for an optimal wavelet kernel that can be used for a specific image processing task. In this paper, a linear convolution neural network was employed to seek a wavelet that minimizes errors and maximizes compression efficiency for an image or a defined image pattern such as microcalcifications in mammograms and bone in computed tomography (CT) head images. We have used this method to evaluate the performance of tap-4 wavelets on mammograms, CTs, magnetic resonance images, and Lena images. We found that the Daubechies wavelet or those wavelets with similar filtering characteristics can produce the highest compression efficiency with the smallest mean-square-error for many image patterns including general image textures as well as microcalcifications in digital mammograms. However, the Haar wavelet produces the best results on sharp edges and low-noise smooth areas. We also found that a special wavelet whose low-pass filter coefficients are 0.32252136, 0.85258927, 1.38458542, and -0.14548269) produces the best preservation outcomes in all tested microcalcification features including the peak signal-to-noise ratio, the contrast and the figure of merit in the wavelet lossy compression scheme. Having analyzed the spectrum of the wavelet filters, we can find the compression outcomes and feature preservation characteristics as a function of wavelets. This newly developed optimization approach can be generalized to other image analysis applications where a wavelet decomposition is employed.

  10. Hdr Imaging for Feature Detection on Detailed Architectural Scenes

    Kontogianni, G.; Stathopoulou, E. K.; Georgopoulos, A.; Doulamis, A.


    3D reconstruction relies on accurate detection, extraction, description and matching of image features. This is even truer for complex architectural scenes that pose needs for 3D models of high quality, without any loss of detail in geometry or color. Illumination conditions influence the radiometric quality of images, as standard sensors cannot depict properly a wide range of intensities in the same scene. Indeed, overexposed or underexposed pixels cause irreplaceable information loss and degrade digital representation. Images taken under extreme lighting environments may be thus prohibitive for feature detection/extraction and consequently for matching and 3D reconstruction. High Dynamic Range (HDR) images could be helpful for these operators because they broaden the limits of illumination range that Standard or Low Dynamic Range (SDR/LDR) images can capture and increase in this way the amount of details contained in the image. Experimental results of this study prove this assumption as they examine state of the art feature detectors applied both on standard dynamic range and HDR images.


    G. Kontogianni


    Full Text Available 3D reconstruction relies on accurate detection, extraction, description and matching of image features. This is even truer for complex architectural scenes that pose needs for 3D models of high quality, without any loss of detail in geometry or color. Illumination conditions influence the radiometric quality of images, as standard sensors cannot depict properly a wide range of intensities in the same scene. Indeed, overexposed or underexposed pixels cause irreplaceable information loss and degrade digital representation. Images taken under extreme lighting environments may be thus prohibitive for feature detection/extraction and consequently for matching and 3D reconstruction. High Dynamic Range (HDR images could be helpful for these operators because they broaden the limits of illumination range that Standard or Low Dynamic Range (SDR/LDR images can capture and increase in this way the amount of details contained in the image. Experimental results of this study prove this assumption as they examine state of the art feature detectors applied both on standard dynamic range and HDR images.

  12. Content Based Image Recognition by Information Fusion with Multiview Features

    Rik Das


    Full Text Available Substantial research interest has been observed in the field of object recognition as a vital component for modern intelligent systems. Content based image classification and retrieval have been considered as two popular techniques for identifying the object of interest. Feature extraction has played the pivotal role towards successful implementation of the aforesaid techniques. The paper has presented two novel techniques of feature extraction from diverse image categories both in spatial domain and in frequency domain. The multi view features from the image categories were evaluated for classification and retrieval performances by means of a fusion based recognition architecture. The experimentation was carried out with four different popular public datasets. The proposed fusion framework has exhibited an average increase of 24.71% and 20.78% in precision rates for classification and retrieval respectively, when compared to state-of-the art techniques. The experimental findings were validated with a paired t test for statistical significance.

  13. Automated Classification of Glaucoma Images by Wavelet Energy Features



    Full Text Available Glaucoma is the second leading cause of blindness worldwide. As glaucoma progresses, more optic nerve tissue is lost and the optic cup grows which leads to vision loss. This paper compiles a systemthat could be used by non-experts to filtrate cases of patients not affected by the disease. This work proposes glaucomatous image classification using texture features within images and efficient glaucoma classification based on Probabilistic Neural Network (PNN. Energy distribution over wavelet sub bands is applied to compute these texture features. Wavelet features were obtained from the daubechies (db3, symlets (sym3, and biorthogonal (bio3.3, bio3.5, and bio3.7 wavelet filters. It uses a technique to extract energy signatures obtained using 2-D discrete wavelet transform and the energy obtained from the detailed coefficients can be used to distinguish between normal and glaucomatous images. We observedan accuracy of around 95%, this demonstrates the effectiveness of these methods.

  14. Feature element theory for image recognition and retrieval

    Xu, Yin; Zhang, Yujin


    Traditional systems of image retrieval work as black boxes. The concrete process and result data or coefficients are not the real care point. Thus it brings the problem: these systems cannot fulfill general semantic application. To overcome the problem we turn to psychology and neuroscience to study the cognition mechanism of human brain. Based on the analysis of experiments and evidences, a new hypothesis - element presence theory is proposed to explain the truth of the whole visual cognition. As its basic level that deals with low-level feature data from camera or retina, feature element theory is illustrated in details. Besides, the evaluation on feature elements is discussed and the illustration on feature element theory based image retrieval system is also given.

  15. Feature extraction with LIDAR data and aerial images

    Mao, Jianhua; Liu, Yanjing; Cheng, Penggen; Li, Xianhua; Zeng, Qihong; Xia, Jing


    Raw LIDAR data is a irregular spacing 3D point cloud including reflections from bare ground, buildings, vegetation and vehicles etc., and the first task of the data analyses of point cloud is feature extraction. However, the interpretability of LIDAR point cloud is often limited due to the fact that no object information is provided, and the complex earth topography and object morphology make it impossible for a single operator to classify all the point cloud precisely 100%. In this paper, a hierarchy method for feature extraction with LIDAR data and aerial images is discussed. The aerial images provide us information of objects figuration and spatial distribution, and hierarchic classification of features makes it easy to apply automatic filters progressively. And the experiment results show that, using this method, it was possible to detect more object information and get a better result of feature extraction than using automatic filters alone.

  16. Analysis of Cry Features in Newborns with Differential Fetal Growth.

    Zeskind, Philip Sanford; Ramey, Craig T.


    Describes the relationship between neonatal crying and anthropometric indices of fetal growth. No differences were found between cry features of underweight and overweight infants; both groups required more stimulation than average weight infants to elicit crying. It is suggested that certain cry features may reflect the risk status of neonates…

  17. Imaging features of complex sclerosing lesions of the breast

    Myong, Joo Hwa; Choi, Byung Gil; Kim, Sung Hun; Kang, Bong Joo; Lee, Ah Won; Song, Byung Joo [Seoul St. Mary' s Hospital, The Catholic University of Korea College of Medicine, Seoul (Korea, Republic of)


    The purpose of this study was to evaluate the imaging features of complex sclerosing lesions of the breast and to assess the rate of upgrade to breast cancer. From March 2008 to May 2012, seven lesions were confirmed as complex sclerosing lesions by ultrasonography-guided core needle biopsy. Final results by either surgical excision or follow-up imaging studies were reviewed to assess the rate of upgrade to breast cancer. Two radiologists retrospectively analyzed the imaging findings according to the Breast Imaging Reporting and Data System classification. Five lesions underwent subsequent surgical excision and two of them revealed ductal carcinoma in situ (n=1) and invasive ductal carcinoma (n=1). Our study showed a breast cancer upgrade rate of 28.6% (2 of 7 lesions). Two lesions were stable on imaging follow-up beyond 1 year. The mammographic features included masses (n=4, 57.1%), architectural distortion (n=2, 28.6%), and focal asymmetry (n=1, 14.3%). Common B-mode ultrasonographic features were irregular shape (n=6, 85.7%), spiculated margin (n=5, 71.4 %), and hypoechogenicity (n=7, 100%). The final assessment categories were category 4 (n=6, 85.7%) and category 5 (n=1, 14.3%). The complex sclerosing lesions were commonly mass-like on mammography and showed the suspicious ultrasonographic features of category 4. Due to a high underestimation rate, all complex sclerosing lesions by core needle biopsy should be excised.

  18. Dermoscopy analysis of RGB-images based on comparative features

    Myakinin, Oleg O.; Zakharov, Valery P.; Bratchenko, Ivan A.; Artemyev, Dmitry N.; Neretin, Evgeny Y.; Kozlov, Sergey V.


    In this paper, we propose an algorithm for color and texture analysis for dermoscopic images of human skin based on Haar wavelets, Local Binary Patterns (LBP) and Histogram Analysis. This approach is a modification of «7-point checklist» clinical method. Thus, that is an "absolute" diagnostic method because one is using only features extracted from tumor's ROI (Region of Interest), which can be selected manually and/or using a special algorithm. We propose additional features extracted from the same image for comparative analysis of tumor and healthy skin. We used Euclidean distance, Cosine similarity, and Tanimoto coefficient as comparison metrics between color and texture features extracted from tumor's and healthy skin's ROI separately. A classifier for separating melanoma images from other tumors has been built by SVM (Support Vector Machine) algorithm. Classification's errors with and without comparative features between skin and tumor have been analyzed. Significant increase of recognition quality with comparative features has been demonstrated. Moreover, we analyzed two modes (manual and automatic) for ROI selecting on tumor and healthy skin areas. We have reached 91% of sensitivity using comparative features in contrast with 77% of sensitivity using the only "absolute" method. The specificity was the invariable (94%) in both cases.

  19. Modality prediction of biomedical literature images using multimodal feature representation

    Pelka, Obioma


    Full Text Available This paper presents the modelling approaches performed to automatically predict the modality of images found in biomedical literature. Various state-of-the-art visual features such as Bag-of-Keypoints computed with dense SIFT descriptors, texture features and Joint Composite Descriptors were used for visual image representation. Text representation was obtained by vector quantisation on a Bag-of-Words dictionary generated using attribute importance derived from a χ-test. Computing the principal components separately on each feature, dimension reduction as well as computational load reduction was achieved. Various multiple feature fusions were adopted to supplement visual image information with corresponding text information. The improvement obtained when using multimodal features vs. visual or text features was detected, analysed and evaluated. Random Forest models with 100 to 500 deep trees grown by resampling, a multi class linear kernel SVM with C=0.05 and a late fusion of the two classifiers were used for modality prediction. A Random Forest classifier achieved a higher accuracy and computed Bag-of-Keypoints with dense SIFT descriptors proved to be a better approach than with Lowe SIFT.

  20. Image Retrieval via Relevance Vector Machine with Multiple Features

    Zemin Liu


    Full Text Available With the fast development of computer network technique, there is large amount of image information every day. Researchers have paid more and more attention to the problem of how users quickly retrieving and identifying the images that they may interest. Meanwhile, with the rapid development of artificial intelligence and pattern recognition techniques, it provides people with new thought on the study on complex image retrieval while it’s very difficult for traditional machine learning method to get ideal retrieval results. For this reason, we in this paper propose a new approach for image retrieval based on multiple types of image features and relevance vector machine (RVM. The proposed method, termed as MF-RVM, integrates the informative cures of features and the discrimination ability of RVM. The retrieval experiment is conducted on COREL image library which is collected from internet. The experimental results show that the proposed method can significantly improve the performance for image retrieval, so MF-RVM presented in this paper has very high practicability in image retrieval.

  1. Geometrically invariant color image watermarking scheme using feature points

    WANG XiangYang; MENG Lan; YANG HongYing


    Geometric distortion is known as one of the most difficult attacks to resist.Geometric distortion desynchronizes the location of the watermark and hence causes incorrect watermark detection.In this paper,we propose a geometrically invariant digital watermarking method for color images.In order to synchronize the location for watermark insertion and detection,we use a multi-scale Harris-Laplace detector,by which feature points of a color image can be extracted that are invariant to geometric distortions.Then,the self-adaptive local image region (LIR) detection based on the feature scale theory was considered for watermarking.At each local image region,the watermark is embedded after image normalization.By binding digital watermark with invariant image regions,resilience against geometric distortion can be readily obtained.Our method belongs to the category of blind watermarking techniques,because we do not need the original image during detection.Experimental results show that the proposed color image watermarking is not only invisible and robust against common signal processing such as sharpening,noise adding,and JPEG compression,but also robust against the geometric distortions such as rotation,translation,scaling,row or column removal,shearing,and local random bend.

  2. The fuzzy Hough Transform-feature extraction in medical images

    Philip, K.P.; Dove, E.L.; Stanford, W.; Chandran, K.B. (Univ. of Iowa, Iowa City, IA (United States)); McPherson, D.D.; Gotteiner, N.L. (Northwestern Univ., Chicago, IL (United States). Dept. of Internal Medicine)


    Identification of anatomical features is a necessary step for medical image analysis. Automatic methods for feature identification using conventional pattern recognition techniques typically classify an object as a member of a predefined class of objects, but do not attempt to recover the exact or approximate shape of that object. For this reason, such techniques are usually not sufficient to identify the borders of organs when individual geometry varies in local detail, even though the general geometrical shape is similar. The authors present an algorithm that detects features in an image based on approximate geometrical models. The algorithm is based on the traditional and generalized Hough Transforms but includes notions from fuzzy set theory. The authors use the new algorithm to roughly estimate the actual locations of boundaries of an internal organ, and from this estimate, to determine a region of interest around the organ. Based on this rough estimate of the border location, and the derived region of interest, the authors find the final estimate of the true borders with other image processing techniques. The authors present results that demonstrate that the algorithm was successfully used to estimate the approximate location of the chest wall in humans, and of the left ventricular contours of a dog heart obtained from cine-computed tomographic images. The authors use this fuzzy Hough Transform algorithm as part of a larger procedures to automatically identify the myocardial contours of the heart. This algorithm may also allow for more rapid image processing and clinical decision making in other medical imaging applications.

  3. Feature analysis for detecting people from remotely sensed images

    Sirmacek, Beril; Reinartz, Peter


    We propose a novel approach using airborne image sequences for detecting dense crowds and individuals. Although airborne images of this resolution range are not enough to see each person in detail, we can still notice a change of color and intensity components of the acquired image in the location where a person exists. Therefore, we propose a local feature detection-based probabilistic framework to detect people automatically. Extracted local features behave as observations of the probability density function (PDF) of the people locations to be estimated. Using an adaptive kernel density estimation method, we estimate the corresponding PDF. First, we use estimated PDF to detect boundaries of dense crowds. After that, using background information of dense crowds and previously extracted local features, we detect other people in noncrowd regions automatically for each image in the sequence. To test our crowd and people detection algorithm, we use airborne images taken over Munich during the Oktoberfest event, two different open-air concerts, and an outdoor festival. In addition, we apply tests on GeoEye-1 satellite images. Our experimental results indicate possible use of the algorithm in real-life mass events.

  4. Feature statistic analysis of ultrasound images of liver cancer

    Huang, Shuqin; Ding, Mingyue; Zhang, Songgeng


    In this paper, a specific feature analysis of liver ultrasound images including normal liver, liver cancer especially hepatocellular carcinoma (HCC) and other hepatopathy is discussed. According to the classification of hepatocellular carcinoma (HCC), primary carcinoma is divided into four types. 15 features from single gray-level statistic, gray-level co-occurrence matrix (GLCM), and gray-level run-length matrix (GLRLM) are extracted. Experiments for the discrimination of each type of HCC, normal liver, fatty liver, angioma and hepatic abscess have been conducted. Corresponding features to potentially discriminate them are found.

  5. Scene classification of infrared images based on texture feature

    Zhang, Xiao; Bai, Tingzhu; Shang, Fei


    Scene Classification refers to as assigning a physical scene into one of a set of predefined categories. Utilizing the method texture feature is good for providing the approach to classify scenes. Texture can be considered to be repeating patterns of local variation of pixel intensities. And texture analysis is important in many applications of computer image analysis for classification or segmentation of images based on local spatial variations of intensity. Texture describes the structural information of images, so it provides another data to classify comparing to the spectrum. Now, infrared thermal imagers are used in different kinds of fields. Since infrared images of the objects reflect their own thermal radiation, there are some shortcomings of infrared images: the poor contrast between the objectives and background, the effects of blurs edges, much noise and so on. Because of these shortcomings, it is difficult to extract to the texture feature of infrared images. In this paper we have developed an infrared image texture feature-based algorithm to classify scenes of infrared images. This paper researches texture extraction using Gabor wavelet transform. The transformation of Gabor has excellent capability in analysis the frequency and direction of the partial district. Gabor wavelets is chosen for its biological relevance and technical properties In the first place, after introducing the Gabor wavelet transform and the texture analysis methods, the infrared images are extracted texture feature by Gabor wavelet transform. It is utilized the multi-scale property of Gabor filter. In the second place, we take multi-dimensional means and standard deviation with different scales and directions as texture parameters. The last stage is classification of scene texture parameters with least squares support vector machine (LS-SVM) algorithm. SVM is based on the principle of structural risk minimization (SRM). Compared with SVM, LS-SVM has overcome the shortcoming of

  6. Investigation of efficient features for image recognition by neural networks.

    Goltsev, Alexander; Gritsenko, Vladimir


    In the paper, effective and simple features for image recognition (named LiRA-features) are investigated in the task of handwritten digit recognition. Two neural network classifiers are considered-a modified 3-layer perceptron LiRA and a modular assembly neural network. A method of feature selection is proposed that analyses connection weights formed in the preliminary learning process of a neural network classifier. In the experiments using the MNIST database of handwritten digits, the feature selection procedure allows reduction of feature number (from 60 000 to 7000) preserving comparable recognition capability while accelerating computations. Experimental comparison between the LiRA perceptron and the modular assembly neural network is accomplished, which shows that recognition capability of the modular assembly neural network is somewhat better.

  7. Two-level hierarchical feature learning for image classification

    Guang-hui SONG; Xiao-gang JIN; Gen-lang CHEN; Yan NIE


    In some image classifi cation tasks, similarities among different categories are different and the samples are usually misclassifi ed as highly similar categories. To distinguish highly similar categories, more specifi c features are required so that the classifi er can improve the classifi cation performance. In this paper, we propose a novel two-level hierarchical feature learning framework based on the deep convolutional neural network (CNN), which is simple and effective. First, the deep feature extractors of different levels are trained using the transfer learning method that fi ne-tunes the pre-trained deep CNN model toward the new target dataset. Second, the general feature extracted from all the categories and the specifi c feature extracted from highly similar categories are fused into a feature vector. Then the fi nal feature representation is fed into a linear classifi er. Finally, experiments using the Caltech-256, Oxford Flower-102, and Tasmania Coral Point Count (CPC) datasets demonstrate that the expression ability of the deep features resulting from two-level hierarchical feature learning is powerful. Our proposed method effectively increases the classifi cation accuracy in comparison with fl at multiple classifi cation methods.

  8. Computational optical sensing and imaging: introduction to feature issue.

    Gerwe, David R; Harvey, Andrew; Gehm, Michael E


    The 2012 Computational Optical Sensing and Imaging (COSI) conference of the Optical Society of America was one of six colocated meetings composing the Imaging and Applied Optics Congress held in Monterey, California, 24-28 June. COSI, together with the Imaging Systems and Applications, Optical Sensors, Applied Industrial Optics, and Optical Remote Sensing of the Environment conferences, brought together a diverse group of scientists and engineers sharing a common interest in measuring and processing of information carried by optical fields. This special feature includes several papers based on presentations given at the 2012 COSI conference as well as independent contributions, which together highlight several important trends.

  9. Classifying Cyst and Tumor Lesion Using Support Vector Machine Based on Dental Panoramic Images Texture Features

    Nurtanio, Ingrid


    Dental radiographs are essential in diagnosing the pathology of the jaw. However, similar radiographic appearance of jaw lesions causes difficulties in differentiating cyst from tumor. Therefore, we conducted a development of computer-aided classification system for cyst and tumor lesions in dental panoramic images. The proposed system consists of feature extraction based on texture using the first-order statistics texture (FO), Gray Level Co-occurrence Matrix (GLCM) and Gray Level Run ...

  10. Deep optical images of Malin 1 reveal new features

    Galaz, Gaspar; Suc, Vincent; Busta, Luis; Lizana, Guadalupe; Infante, Leopoldo; Royo, Santiago


    We present Megacam deep optical images (g and r) of Malin 1 obtained with the 6.5m Magellan/Clay telescope, detecting structures down to ~ 28 B mag arcsec-2. In order to enhance galaxy features buried in the noise, we use a noise reduction filter based on the total generalized variation regularizator. This method allows us to detect and resolve very faint morphological features, including spiral arms, with a high visual contrast. For the first time, we can appreciate an optical image of Malin 1 and its morphology in full view. The images provide unprecedented detail, compared to those obtained in the past with photographic plates and CCD, including HST imaging. We detect two peculiar features in the disk/spiral arms. The analysis suggests that the first one is possibly a background galaxy, and the second is an apparent stream without a clear nature, but could be related to the claimed past interaction between Malin 1 and the galaxy SDSSJ123708.91 + 142253.2. Malin 1 exhibits features suggesting the presence o...

  11. Identification and Quantification Soil Redoximorphic Features by Digital Image Processing

    Soil redoximorphic features (SRFs) have provided scientists and land managers with insight into relative soil moisture for approximately 60 years. The overall objective of this study was to develop a new method of SRF identification and quantification from soil cores using a digital camera and imag...

  12. MR Imaging Features of Obturator Internus Bursa of the Hip

    Hwang, Ji Young; Lee, Sun Wha; Kim, Jong Oh [School of Medicine, Ewha Womans University, Seoul (Korea, Republic of)


    The authors report two cases with distension of the obturator internus bursa identified on MR images, and describe the location and characteristic features of obturator internus bursitis; the 'boomerang'-shaped fluid distension between the obturator internus tendon and the posterior grooved surface of the ischium

  13. Multi-modal image registration using structural features.

    Kasiri, Keyvan; Clausi, David A; Fieguth, Paul


    Multi-modal image registration has been a challenging task in medical images because of the complex intensity relationship between images to be aligned. Registration methods often rely on the statistical intensity relationship between the images which suffers from problems such as statistical insufficiency. The proposed registration method works based on extracting structural features by utilizing the complex phase and gradient-based information. By employing structural relationships between different modalities instead of complex similarity measures, the multi-modal registration problem is converted into a mono-modal one. Therefore, conventional mono-modal similarity measures can be utilized to evaluate the registration results. This new registration paradigm has been tested on magnetic resonance (MR) brain images of different modes. The method has been evaluated based on target registration error (TRE) to determine alignment accuracy. Quantitative results demonstrate that the proposed method is capable of achieving comparable registration accuracy compared to the conventional mutual information.

  14. Moment feature based fast feature extraction algorithm for moving object detection using aerial images.

    A F M Saifuddin Saif

    Full Text Available Fast and computationally less complex feature extraction for moving object detection using aerial images from unmanned aerial vehicles (UAVs remains as an elusive goal in the field of computer vision research. The types of features used in current studies concerning moving object detection are typically chosen based on improving detection rate rather than on providing fast and computationally less complex feature extraction methods. Because moving object detection using aerial images from UAVs involves motion as seen from a certain altitude, effective and fast feature extraction is a vital issue for optimum detection performance. This research proposes a two-layer bucket approach based on a new feature extraction algorithm referred to as the moment-based feature extraction algorithm (MFEA. Because a moment represents the coherent intensity of pixels and motion estimation is a motion pixel intensity measurement, this research used this relation to develop the proposed algorithm. The experimental results reveal the successful performance of the proposed MFEA algorithm and the proposed methodology.

  15. No-reference face image assessment based on deep features

    Liu, Guirong; Xu, Yi; Lan, Jinpeng


    Face quality assessment is important to improve the performance of face recognition system. For instance, it is required to select images of good quality to improve recognition rate for the person of interest. Current methods mostly depend on traditional image assessment, which use prior knowledge of human vision system. As a result, the quality score of face images shows consistency with human vision perception but deviates from the processing procedure of a real face recognition system. It is the fact that the state-of-art face recognition systems are all built on deep neural networks. Naturally, it is expected to propose an efficient quality scoring method of face images, which should show high consistency with the recognition rate of face images from current face recognition systems. This paper proposes a non-reference face image assessment algorithm based on the deep features, which is capable of predicting the recognition rate of face images. The proposed face image assessment algorithm provides a promising tool to filter out the good input images for the real face recognition system to achieve high recognition rate.

  16. New learning subspace method for image feature extraction

    CAO Jian-hai; LI Long; LU Chang-hou


    A new method of Windows Minimum/Maximum Module Learning Subspace Algorithm(WMMLSA) for image feature extraction is presented. The WMMLSM is insensitive to the order of the training samples and can regulate effectively the radical vectors of an image feature subspace through selecting the study samples for subspace iterative learning algorithm,so it can improve the robustness and generalization capacity of a pattern subspace and enhance the recognition rate of a classifier. At the same time,a pattern subspace is built by the PCA method. The classifier based on WMMLSM is successfully applied to recognize the pressed characters on the gray-scale images. The results indicate that the correct recognition rate on WMMLSM is higher than that on Average Learning Subspace Method,and that the training speed and the classification speed are both improved. The new method is more applicable and efficient.

  17. Characterisation of Feature Points in Eye Fundus Images

    Calvo, D.; Ortega, M.; Penedo, M. G.; Rouco, J.

    The retinal vessel tree adds decisive knowledge in the diagnosis of numerous opthalmologic pathologies such as hypertension or diabetes. One of the problems in the analysis of the retinal vessel tree is the lack of information in terms of vessels depth as the image acquisition usually leads to a 2D image. This situation provokes a scenario where two different vessels coinciding in a point could be interpreted as a vessel forking into a bifurcation. That is why, for traking and labelling the retinal vascular tree, bifurcations and crossovers of vessels are considered feature points. In this work a novel method for these retinal vessel tree feature points detection and classification is introduced. The method applies image techniques such as filters or thinning to obtain the adequate structure to detect the points and sets a classification of these points studying its environment. The methodology is tested using a standard database and the results show high classification capabilities.


    Soumen Bag


    Full Text Available Feature selection and extraction plays an important role in different classification based problems such as face recognition, signature verification, optical character recognition (OCR etc. The performance of OCR highly depends on the proper selection and extraction of feature set. In this paper, we present novel features based on the topography of a character as visible from different viewing directions on a 2D plane. By topography of a character we mean the structural features of the strokes and their spatial relations. In this work we develop topographic features of strokes visible with respect to views from different directions (e.g. North, South, East, and West. We consider three types of topographic features: closed region, convexity of strokes, and straight line strokes. These features are represented as a shapebased graph which acts as an invariant feature set for discriminating very similar type characters efficiently. We have tested the proposed method on printed and handwritten Bengali and Hindi character images. Initial results demonstrate the efficacy of our approach.

  19. Topographic Feature Extraction for Bengali and Hindi Character Images

    Soumen Bag


    Full Text Available Feature selection and extraction plays an important role in different classification based problems such as face recognition, signature verification, optical character recognition (OCR etc. The performance of OCR highly depends on the proper selection and extraction of feature set. In this paper, we present novel features based on the topography of a character as visible from different viewing directions on a 2D plane. By topography of a character we mean the structural features of the strokes and their spatial relations. In this work we develop topographic features of strokes visible with respect to views from different directions (e.g. North, South, East, and West. We consider three types of topographic features: closed region, convexity of strokes, and straight line strokes. These features are represented as a shapebased graph which acts as an invariant feature set for discriminating very similar type characters efficiently. We have tested the proposed method on printed and handwritten Bengali and Hindi character images. Initial results demonstrate the efficacy of our approach.

  20. Image Recognition and Feature Detection in Solar Physics

    Martens, Petrus C.


    The Solar Dynamics Observatory (SDO) data repository will dwarf the archives of all previous solar physics missions put together. NASA recognized early on that the traditional methods of analyzing the data -- solar scientists and grad students in particular analyzing the images by hand -- would simply not work and tasked our Feature Finding Team (FFT) with developing automated feature recognition modules for solar events and phenomena likely to be observed by SDO. Having these metadata available on-line will enable solar scientist to conduct statistical studies involving large sets of events that would be impossible now with traditional means. We have followed a two-track approach in our project: we have been developing some existing task-specific solar feature finding modules to be "pipe-line" ready for the stream of SDO data, plus we are designing a few new modules. Secondly, we took it upon us to develop an entirely new "trainable" module that would be capable of identifying different types of solar phenomena starting from a limited number of user-provided examples. Both approaches are now reaching fruition, and I will show examples and movies with results from several of our feature finding modules. In the second part of my presentation I will focus on our “trainable” module, which is the most innovative in character. First, there is the strong similarity between solar and medical X-ray images with regard to their texture, which has allowed us to apply some advances made in medical image recognition. Second, we have found that there is a strong similarity between the way our trainable module works and the way our brain recognizes images. The brain can quickly recognize similar images from key characteristics, just as our code does. We conclude from that that our approach represents the beginning of a more human-like procedure for computer image recognition.

  1. Single Image Superresolution via Directional Group Sparsity and Directional Features.

    Li, Xiaoyan; He, Hongjie; Wang, Ruxin; Tao, Dacheng


    Single image superresolution (SR) aims to construct a high-resolution version from a single low-resolution (LR) image. The SR reconstruction is challenging because of the missing details in the given LR image. Thus, it is critical to explore and exploit effective prior knowledge for boosting the reconstruction performance. In this paper, we propose a novel SR method by exploiting both the directional group sparsity of the image gradients and the directional features in similarity weight estimation. The proposed SR approach is based on two observations: 1) most of the sharp edges are oriented in a limited number of directions and 2) an image pixel can be estimated by the weighted averaging of its neighbors. In consideration of these observations, we apply the curvelet transform to extract directional features which are then used for region selection and weight estimation. A combined total variation regularizer is presented which assumes that the gradients in natural images have a straightforward group sparsity structure. In addition, a directional nonlocal means regularization term takes pixel values and directional information into account to suppress unwanted artifacts. By assembling the designed regularization terms, we solve the SR problem of an energy function with minimal reconstruction error by applying a framework of templates for first-order conic solvers. The thorough quantitative and qualitative results in terms of peak signal-to-noise ratio, structural similarity, information fidelity criterion, and preference matrix demonstrate that the proposed approach achieves higher quality SR reconstruction than the state-of-the-art algorithms.

  2. Document image retrieval based on multi-density features

    HU Zhilan; LIN Xinggang; YAN Hong


    The development of document image databases is becoming a challenge for document image retrieval techniques.Traditional layout-reconstructed-based methods rely on high quality document images as well as an optical character recognition (OCR) precision,and can only deal with several widely used languages.The complexity of document layouts greatly hinders layout analysis-based approaches.This paper describes a multi-density feature based algorithm for binary document images,which is independent of OCR or layout analyses.The text area was extracted after preprocessing such as skew correction and marginal noise removal.Then the aspect ratio and multi-density features were extracted from the text area to select the best candidates from the document image database.Experimental results show that this approach is simple with loss rates less than 3% and can efficiently analyze images with different resolutions and different input systems.The system is also robust to noise due to its notes and complex layouts,etc.

  3. GPU Accelerated Automated Feature Extraction From Satellite Images

    K. Phani Tejaswi


    Full Text Available The availability of large volumes of remote sensing data insists on higher degree of automation in featureextraction, making it a need of thehour. Fusingdata from multiple sources, such as panchromatic,hyperspectraland LiDAR sensors, enhances the probability of identifying and extracting features such asbuildings, vegetation or bodies of water by using a combination of spectral and elevation characteristics.Utilizing theaforementioned featuresin remote sensing is impracticable in the absence ofautomation.Whileefforts are underway to reduce human intervention in data processing, this attempt alone may notsuffice. Thehuge quantum of data that needs to be processed entailsaccelerated processing to be enabled.GPUs, which were originally designed to provide efficient visualization,arebeing massively employed forcomputation intensive parallel processing environments. Image processing in general and hence automatedfeatureextraction, is highly computation intensive, where performance improvements have a direct impacton societal needs. In this context, an algorithm has been formulated for automated feature extraction froma panchromatic or multispectral image based on image processing techniques.Two Laplacian of Guassian(LoGmasks were applied on the image individually followed by detection of zero crossing points andextracting the pixels based on their standard deviationwiththe surrounding pixels. The two extractedimages with different LoG masks were combined together which resulted in an image withthe extractedfeatures and edges.Finally the user is at liberty to apply the image smoothing step depending on the noisecontent in the extracted image.The image ispassed through a hybrid median filter toremove the salt andpepper noise from the image.This paper discusses theaforesaidalgorithmforautomated featureextraction, necessity of deployment of GPUs for thesame;system-level challenges and quantifies thebenefits of integrating GPUs in such environment. The

  4. Weighted feature fusion for content-based image retrieval

    Soysal, Omurhan A.; Sumer, Emre


    The feature descriptors such as SIFT (Scale Invariant Feature Transform), SURF (Speeded-up Robust Features) and ORB (Oriented FAST and Rotated BRIEF) are known as the most commonly used solutions for the content-based image retrieval problems. In this paper, a novel approach called "Weighted Feature Fusion" is proposed as a generic solution instead of applying problem-specific descriptors alone. Experiments were performed on two basic data sets of the Inria in order to improve the precision of retrieval results. It was found that in cases where the descriptors were used alone the proposed approach yielded 10-30% more accurate results than the ORB alone. Besides, it yielded 9-22% and 12-29% less False Positives compared to the SIFT alone and SURF alone, respectively.

  5. A flower image retrieval method based on ROI feature

    洪安祥; 陈刚; 李均利; 池哲儒; 张亶


    Flower image retrieval is a very important step for computer-aided plant species recognition. In this paper, we propose an efficient segmentation method based on color clustering and domain knowledge to extract flower regions from flower images. For flower retrieval, we use the color histogram of a flower region to characterize the color features of flower and two shape-based features sets, Centroid-Contour Distance (CCD) and Angle Code Histogram (ACH), to characterize the shape features of a flower contour. Experimental results showed that our flower region extraction method based on color clustering and domain knowledge can produce accurate flower regions. Flower retrieval results on a database of 885 flower images collected from 14 plant species showed that our Region-of-Interest (ROI) based retrieval approach using both color and shape features can perform better than a method based on the global color histogram proposed by Swain and Ballard (1991) and a method based on domain knowledge-driven segmentation and color names proposed by Das et al.(1999).

  6. A flower image retrieval method based on ROI feature

    洪安祥; 陈刚; 李均利; 池哲儒; 张亶


    Flower image retrieval is a very important step for computer-aided plant species recognition.In this paper,we propose an efficient segmentation method based on color clustering and domain knowledge to extract flower regions from flower images.For flower retrieval,we use the color histogram of a flower region to characterize the color features of flower and two shape-based features sets,Centroid-Contour Distance(CCD)and Angle Code Histogram(ACH),to characterize the shape features of a flower contour.Experimental results showed that our flower region extraction method based on color clustering and domain knowledge can produce accurate flower regions.Flower retrieval results on a database of 885 flower images collected from 14 plant species showed that our Region-of-Interest(ROD based retrieval approach using both color and shape features can perform better than a method based on the global color histogram proposed by Swain and Ballard(1991)and a method based on domain knowledge-driven segmentation and color names proposed by Das et al.(1999).

  7. Differentiation of digital tb images using texture analysis and rbf classifier.

    Priya, E; Srinivasan, S; Ramakrishnan, S


    In this work, differentiation of positive and negative images of Tuberculosis (TB) sputum smear has been attempted using statistical method based on Gray Level Co-occurrence Matrix (GLCM). The sputum smear images (N=100) recorded under standard image acquisition protocol are considered for this work. Second order statistical texture analysis is performed on the acquired images using GLCM method and a set of nineteen features are derived. Principal Component Analysis (PCA) is then employed to reduce feature sets, to enhance the efficiency of differentiation and to reduce the redundancy. These feature sets are further classified using Radial Basis Function (RBF) classifier. Results show that GLCM is able to differentiate positive and negative TB images. Correlation is found to be high for many of the parameters. Application of PCA reduced the number of features to four which had maximum magnitude in the first principal component. Higher classification accuracy is achieved using RBF classifier. It appears that this method of texture analysis could be useful to develop automated system for characterization and classification of digital TB sputum smear images.

  8. Estimating perception of scene layout properties from global image features.

    Ross, Michael G; Oliva, Aude


    The relationship between image features and scene structure is central to the study of human visual perception and computer vision, but many of the specifics of real-world layout perception remain unknown. We do not know which image features are relevant to perceiving layout properties, or whether those features provide the same information for every type of image. Furthermore, we do not know the spatial resolutions required for perceiving different properties. This paper describes an experiment and a computational model that provides new insights on these issues. Humans perceive the global spatial layout properties such as dominant depth, openness, and perspective, from a single image. This work describes an algorithm that reliably predicts human layout judgments. This model's predictions are general, not specific to the observers it trained on. Analysis reveals that the optimal spatial resolutions for determining layout vary with the content of the space and the property being estimated. Openness is best estimated at high resolution, depth is best estimated at medium resolution, and perspective is best estimated at low resolution. Given the reliability and simplicity of estimating the global layout of real-world environments, this model could help resolve perceptual ambiguities encountered by more detailed scene reconstruction schemas.

  9. Central nervous system lymphoma: magnetic resonance imaging features at presentation.

    Schwingel, Ricardo; Reis, Fabiano; Zanardi, Veronica A; Queiroz, Luciano S; França, Marcondes C


    This paper aimed at studying presentations of the central nervous system (CNS) lymphoma using structural images obtained by magnetic resonance imaging (MRI). The MRI features at presentation of 15 patients diagnosed with CNS lymphoma in a university hospital, between January 1999 and March 2011, were analyzed by frequency and cross tabulation. All patients had supratentorial lesions; and four had infra- and supratentorial lesions. The signal intensity on T1 and T2 weighted images was predominantly hypo- or isointense. In the T2 weighted images, single lesions were associated with a hypointense signal component. Six patients presented necrosis, all of them showed perilesional abnormal white matter, nine had meningeal involvement, and five had subependymal spread. Subependymal spread and meningeal involvement tended to occur in younger patients. Presentations of lymphoma are very pleomorphic, but some of them should point to this diagnostic possibility.

  10. Feature Based Correspondence: A Comparative Study on Image Matching Algorithms

    Munim Tanvir


    Full Text Available Image matching and recognition are the crux of computer vision and have a major part to play in everyday lives. From industrial robots to surveillance cameras, from autonomous vehicles to medical imaging and from missile guidance to space exploration vehicles computer vision and hence image matching is embedded in our lives. This communication presents a comparative study on the prevalent matching algorithms, addressing their restrictions and providing a criterion to define the level of efficiency likely to be expected from an algorithm. The study includes the feature detection and matching techniques used by these prevalent algorithms to allow a deeper insight. The chief aim of the study is to deliver a source of comprehensive reference for the researchers involved in image matching, regardless of specific applications.

  11. Adaptive feature-specific imaging: a face recognition example.

    Baheti, Pawan K; Neifeld, Mark A


    We present an adaptive feature-specific imaging (AFSI) system and consider its application to a face recognition task. The proposed system makes use of previous measurements to adapt the projection basis at each step. Using sequential hypothesis testing, we compare AFSI with static-FSI (SFSI) and static or adaptive conventional imaging in terms of the number of measurements required to achieve a specified probability of misclassification (Pe). The AFSI system exhibits significant improvement compared to SFSI and conventional imaging at low signal-to-noise ratio (SNR). It is shown that for M=4 hypotheses and desired Pe=10(-2), AFSI requires 100 times fewer measurements than the adaptive conventional imager at SNR= -20 dB. We also show a trade-off, in terms of average detection time, between measurement SNR and adaptation advantage, resulting in an optimal value of integration time (equivalent to SNR) per measurement.

  12. Differential Diagnosis of Patients with Inconclusive Parkinsonian Features Using [{sup 18}F]FP-CIT PET/CT

    Park, Eunkyung; Hwang, Yu Mi; Lee, Channyoung; Oh, Sun Young; Kim, Young Chul; Choe, Jae Gol; Park, Kun Woo [Korea Univ., Seoul (Korea, Republic of); Kim, Sujin [Seoul National Univ. College of Medicine, Seoul (Korea, Republic of)


    It is often difficult to differentiate parkinsonism, especially when patients show uncertain parkinsonian features. We investigated the usefulness of dopamine transporter (DAT) imaging for the differential diagnosis of inconclusive parkinsonism using [{sup 18}F]FP-CIT PET. Twenty-four patients with inconclusive parkinsonian features at initial clinical evaluation and nine healthy controls were studied. Patients consisted of three subgroups: nine patients whose diagnoses were unclear concerning whether they had idiopathic Parkinson's disease or drug-induced parkinsonism ('PD/DIP'), nine patients who fulfilled neither the diagnostic criteria of PD nor of essential tremor ('PD/ET'), and six patients who were alleged to have either PD or atypical parkinsonian syndrome ('PD/APS'). Brain PET images were obtained 120 min after injection of 185 MBq [{sup 18}F]FP-CIT. Imaging results were quantified and compared with follow-up clinical diagnoses. Overall, 11 of 24 patients demonstrated abnormally decreased DAT availability on the PET scans, whereas 13 were normal. PET results could diagnose PD/DIP and PD/ET patients as having PD in six patients, DIP in seven, and ET in five; however, the diagnoses of all six PD/APS patients remained inconclusive. Among 15 patients who obtained a final follow-up diagnosis, the image-based diagnosis was congruent with the follow-up diagnosis in 11 patients. Four unsolved cases had normal DAT availability, but clinically progressed to PD during the follow-up period. [{sup 18}F]FP-CIT PET imaging is useful in the differential diagnosis of patients with inconclusive parkinsonian features, except in patients who show atypical features or who eventually progress to PD.

  13. Mucinous cystic neoplasms of the pancreas: Imaging features and diagnostic difficulties

    Scott, J.; Martin, I.; Redhead, D.; Hammond, P.; Garden, O.J


    AIMS: To review the imaging features of mucinous cystic neoplasms (MCNs) of the pancreas and to highlight difficulties in differentiating these lesions from pancreatic pseudocysts. MATERIALS AND METHODS: The imaging investigations, case notes and histopathology of 13 patients who underwent surgery for an MCN of the pancreas, were reviewed. RESULTS: An erroneous diagnosis of a pancreatic pseudocyst had been made in five of the 13 cases and in two patients cystenterostomy had been performed. Only one patient had a documented history of acute pancreatitis although mildly elevated serum amylase levels were identified in a further five cases. CT and US correctly diagnosed a cystic pancreatic mass in all 13 patients, however cross-sectional imaging features of neoplasia, such as septae, cyst wall calcification, focal thickening of the cyst wall and papillary projections, were absent in five (38%) cases. Coexistent imaging features of chronic pancreatitis were present in five of the 13 patients and in six resected specimens. Cyst wall calcification occurred only in malignant lesions and there was no relationship between cyst size and the degree of malignancy. While ERCP, angiography, and percutaneous needle aspiration may provide additional information, the majority of these examinations were either unhelpful or even misleading. CONCLUSION: MCNs of the pancreas are frequently diagnosed and mismanaged as pancreatic pseudocysts with an associated increase in patient morbidity and mortality. Diagnostic imaging can help to distinguish MCNs from pseudocysts when there are features of neoplasia present, however, no imaging investigation can reliably differentiate the two conditions in all cases. If clinical doubt remains, it is preferable to err on the side of safety and either employ a 'wait and watch' strategy or to resect a cystic pancreatic lesion rather than drain a potentially malignant MCN. Scott, J. (2000)

  14. Electronic image stabilization system based on global feature tracking

    Zhu Juanjuan; Guo Baolong


    A new robust electronic image stabilization system is presented, which involves feature-point, tracking based global motion estimation and Kalman filtering based motion compensation. First, global motion is estimated from the local motions of selected feature points. Considering the local moving objects or the inevitable mismatch,the matching validation, based on the stable relative distance between the points set is proposed, thus maintaining high accuracy and robustness. Next, the global motion parameters are accumulated for correction by Kalman filter-ation. The experimental result illustrates that the proposed system is effective to stabilize translational, rotational,and zooming jitter and robust to local motions.

  15. The fuzzy Hough transform-feature extraction in medical images.

    Philip, K P; Dove, E L; McPherson, D D; Gotteiner, N L; Stanford, W; Chandran, K B


    Identification of anatomical features is a necessary step for medical image analysis. Automatic methods for feature identification using conventional pattern recognition techniques typically classify an object as a member of a predefined class of objects, but do not attempt to recover the exact or approximate shape of that object. For this reason, such techniques are usually not sufficient to identify the borders of organs when individual geometry varies in local detail, even though the general geometrical shape is similar. The authors present an algorithm that detects features in an image based on approximate geometrical models. The algorithm is based on the traditional and generalized Hough Transforms but includes notions from fuzzy set theory. The authors use the new algorithm to roughly estimate the actual locations of boundaries of an internal organ, and from this estimate, to determine a region of interest around the organ. Based on this rough estimate of the border location, and the derived region of interest, the authors find the final (improved) estimate of the true borders with other (subsequently used) image processing techniques. They present results that demonstrate that the algorithm was successfully used to estimate the approximate location of the chest wall in humans, and of the left ventricular contours of a dog heart obtained from cine-computed tomographic images. The authors use this fuzzy Hough transform algorithm as part of a larger procedure to automatically identify the myocardial contours of the heart. This algorithm may also allow for more rapid image processing and clinical decision making in other medical imaging applications.

  16. SAR Data Fusion Imaging Method Oriented to Target Feature Extraction

    Yang Wei


    Full Text Available To deal with the difficulty for target outlines extracting precisely due to neglect of target scattering characteristic variation during the processing of high-resolution space-borne SAR data, a novel fusion imaging method is proposed oriented to target feature extraction. Firstly, several important aspects that affect target feature extraction and SAR image quality are analyzed, including curved orbit, stop-and-go approximation, atmospheric delay, and high-order residual phase error. Furthermore, the corresponding compensation methods are addressed as well. Based on the analysis, the mathematical model of SAR echo combined with target space-time spectrum is established for explaining the space-time-frequency change rule of target scattering characteristic. Moreover, a fusion imaging strategy and method under high-resolution and ultra-large observation angle range conditions are put forward to improve SAR quality by fusion processing in range-doppler and image domain. Finally, simulations based on typical military targets are used to verify the effectiveness of the fusion imaging method.

  17. Incorporating global information in feature-based multimodal image registration

    Li, Yong; Stevenson, Robert


    A multimodal image registration framework based on searching the best matched keypoints and the incorporation of global information is proposed. It comprises two key elements: keypoint detection and an iterative process. Keypoints are detected from both the reference and test images. For each test keypoint, a number of reference keypoints are chosen as mapping candidates. A triplet of keypoint mappings determine an affine transformation that is evaluated using a similarity metric between the reference image and the transformed test image by the determined transformation. An iterative process is conducted on triplets of keypoint mappings, keeping track of the best matched reference keypoint. Random sample consensus and mutual information are applied to eliminate outlier keypoint mappings. The similarity metric is defined to be the number of overlapped edge pixels over the entire images, allowing for global information to be incorporated in the evaluation of triplets of mappings. The performance of the framework is investigated with keypoints extracted by scale invariant feature transform and partial intensity invariant feature descriptor. Experimental results show that the proposed framework can provide more accurate registration than existing methods.

  18. Cervical spine injury in the elderly: imaging features

    Ehara, S. [Dept. of Radiology, Iwate Medical University School of Medicine, Morioka (Japan); Shimamura, Tadashi [Dept. of Orthopedic Surgery, Iwate Medical University School of Medicine, Morioka (Japan)


    An increase in the elderly population has resulted in an increased incidence of cervical spine injury in this group. No specific type of cervical spine trauma is seen in the elderly, although dens fractures are reported to be common. Hyperextension injuries due to falling and the resultant central cord syndrome in the mid and lower cervical segments due to decreased elasticity as a result of spondylosis may be also characteristic. The imaging features of cervical spine injury are often modified by associated spondylosis deformans, DISH and other systemic disorders. The value of MR imaging in such cases is emphasized. (orig.)


    Hynek Lauschmann


    Full Text Available The morphology of fatigue fracture surface (caused by constant cycle loading is strictly related to crack growth rate. This relation may be expressed, among other methods, by means of fractal analysis. Fractal dimension as a single numerical value is not sufficient. Two types of fractal feature vectors are discussed: multifractal and multiparametric. For analysis of images, the box-counting method for 3D is applied with respect to the non-homogeneity of dimensions (two in space, one in brightness. Examples of application are shown: images of several fracture surfaces are analyzed and related to crack growth rate.

  20. Iris image enhancement for feature recognition and extraction

    Mabuza, GP


    Full Text Available Gonzalez, R.C. and Woods, R.E. 2002. Digital Image Processing 2nd Edition, Instructor?s manual .Englewood Cliffs, Prentice Hall, pp 17-36. Proen?a, H. and Alexandre, L.A. 2007. Toward Noncooperative Iris Recognition: A classification approach using... for performing such tasks and yielding better accuracy (Gonzalez & Woods, 2002). METHODOLOGY The block diagram in Figure 2 demonstrates the processes followed to achieve the results. Figure 2: Methodology flow chart Iris image enhancement for feature...

  1. Effect of zooming on texture features of ultrasonic images

    Kyriacou Efthyvoulos


    Full Text Available Abstract Background Unstable carotid plaques on subjective, visual, assessment using B-mode ultrasound scanning appear as echolucent and heterogeneous. Although previous studies on computer assisted plaque characterisation have standardised B-mode images for brightness, improving the objective assessment of echolucency, little progress has been made towards standardisation of texture analysis methods, which assess plaque heterogeneity. The aim of the present study was to investigate the influence of image zooming during ultrasound scanning on textural features and to test whether or not resolution standardisation decreases the variability introduced. Methods Eighteen still B-mode images of carotid plaques were zoomed during carotid scanning (zoom factor 1.3 and both images were transferred to a PC and normalised. Using bilinear and bicubic interpolation, the original images were interpolated in a process of simulating off-line zoom using the same interpolation factor. With the aid of the colour-coded image, carotid plaques of the original, zoomed and two resampled images for each case were outlined and histogram, first order and second order statistics were subsequently calculated. Results Most second order statistics (21/25, 84% were significantly (p Conclusion Texture analysis of ultrasonic plaques should be performed under standardised resolution settings; otherwise a resolution normalisation algorithm should be applied.

  2. Multiwavelets domain singular value features for image texture classification



    A new approach based on multiwavelets transformation and singular value decomposition (SVD) is proposed for the classification of image textures. Lower singular values are truncated based on its energy distribution to classify the textures in the presence of additive white Gaussian noise (AWGN). The proposed approach extracts features such as energy, entropy, local homogeneity and max-min ratio from the selected singular values of multiwavelets transformation coefficients of image textures.The classification was carried out using probabilistic neural network (PNN). Performance of the proposed approach was compared with conventional wavelet domain gray level co-occurrence matrix (GLCM) based features, discrete multiwavelets transformation energy based approach, and HMM based approach. Experimental results showed the superiority of the proposed algorithms when compared with existing algorithms.

  3. Design Approach for Content-based Image Retrieval using Gabor-Zernike features

    Abhinav Deshpande


    Full Text Available The process of extraction of different features from an image is known as Content-based Image Retrieval.Color,Texture and Shape are the major features of an image and play a vital role in the representation of an image..In this paper, a novel method is proposed to extract the region of interest(ROI from an image,prior to extraction of salient features of an image.The image is subjected to normalization so that the noise components due to Gaussian or other types of noises which are present in the image are eliminated and thesuccessfull extraction of various features of an image can be accomplished. Gabor Filters are used to extract the texture feature from an image whereas Zernike Moments can be used to extract the shape feature.The combination of Gabor feature and Zernike feature can be combined to extract Gabor-Zernike Features from an image.

  4. Hyperspectral image classification based on NMF Features Selection Method

    Abe, Bolanle T.; Jordaan, J. A.


    Hyperspectral instruments are capable of collecting hundreds of images corresponding to wavelength channels for the same area on the earth surface. Due to the huge number of features (bands) in hyperspectral imagery, land cover classification procedures are computationally expensive and pose a problem known as the curse of dimensionality. In addition, higher correlation among contiguous bands increases the redundancy within the bands. Hence, dimension reduction of hyperspectral data is very crucial so as to obtain good classification accuracy results. This paper presents a new feature selection technique. Non-negative Matrix Factorization (NMF) algorithm is proposed to obtain reduced relevant features in the input domain of each class label. This aimed to reduce classification error and dimensionality of classification challenges. Indiana pines of the Northwest Indiana dataset is used to evaluate the performance of the proposed method through experiments of features selection and classification. The Waikato Environment for Knowledge Analysis (WEKA) data mining framework is selected as a tool to implement the classification using Support Vector Machines and Neural Network. The selected features subsets are subjected to land cover classification to investigate the performance of the classifiers and how the features size affects classification accuracy. Results obtained shows that performances of the classifiers are significant. The study makes a positive contribution to the problems of hyperspectral imagery by exploring NMF, SVMs and NN to improve classification accuracy. The performances of the classifiers are valuable for decision maker to consider tradeoffs in method accuracy versus method complexity.

  5. Feature-Enhanced, Model-Based Sparse Aperture Imaging


    obtain a sharp estimate of the spatial spectrum that exhibits super-resolution. We propose to use the singular value decomposition ( SVD ) of the data...application in a variety of problems, including image reconstruction and restoration [5], wavelet denoising [6], feature selection in machine learning...on the singular value decomposition ( SVD ) to combine multiple samples and the use of second-order cone programming for optimization of the resulting

  6. Spinal cord ischemia: aetiology, clinical syndromes and imaging features

    Weidauer, Stefan [Frankfurt Univ., Sankt Katharinen Hospital Teaching Hospital, Frankfurt am Main (Germany). Dept. of Neurology; Hattingen, Elke; Berkefeld, Joachim [Frankfurt Univ., Frankfurt am Main (Germany). Inst. of Neuroradiology; Nichtweiss, Michael


    The purpose of this study was to analyse MR imaging features and lesion patterns as defined by compromised vascular territories, correlating them to different clinical syndromes and aetiological aspects. In a 19.8-year period, clinical records and magnetic resonance imaging (MRI) features of 55 consecutive patients suffering from spinal cord ischemia were evaluated. Aetiologies of infarcts were arteriosclerosis of the aorta and vertebral arteries (23.6 %), aortic surgery or interventional aneurysm repair (11 %) and aortic and vertebral artery dissection (11 %), and in 23.6 %, aetiology remained unclear. Infarcts occurred in 38.2 % at the cervical and thoracic level, respectively, and 49 % of patients suffered from centromedullar syndrome caused by anterior spinal artery ischemia. MRI disclosed hyperintense pencil-like lesion pattern on T2WI in 98.2 %, cord swelling in 40 %, enhancement on post-contrast T1WI in 42.9 % and always hyperintense signal on diffusion-weighted imaging (DWI) when acquired. The most common clinical feature in spinal cord ischemia is a centromedullar syndrome, and in contrast to anterior spinal artery ischemia, infarcts in the posterior spinal artery territory are rare. The exclusively cervical location of the spinal sulcal artery syndrome seems to be a likely consequence of anterior spinal artery duplication which is observed preferentially here. (orig.)

  7. Photoacoustic imaging features of intraocular tumors: Retinoblastoma and uveal melanoma

    Xu, Guan; Xue, Yafang; Özkurt, Zeynep Gürsel; Slimani, Naziha; Hu, Zizhong; Wang, Xueding; Xia, Kewen; Ma, Teng; Zhou, Qifa; Demirci, Hakan


    The purpose of this study is to examine the capability of photoacoustic (PA) imaging (PAI) in assessing the unique molecular and architectural features in ocular tumors. A real-time PA and ultrasonography (US) parallel imaging system based on a research US platform was developed to examine retinoblastoma in mice in vivo and human retinoblastoma and uveal melanoma ex vivo. PA signals were generated by optical illumination at 720, 750, 800, 850, 900 and 950 nm delivered through a fiber optical bundle. The optical absorption spectra of the tumors were derived from the PA images. The optical absorption spectrum of each tumor was quantified by fitting to a polynomial model. The microscopic architectures of the tumors were quantified by frequency domain analysis of the PA signals. Both the optical spectral and architectural features agree with the histological findings of the tumors. The mouse and human retinoblastoma showed comparable total optical absorption spectra at a correlation of 0.95 (p<0.005). The quantitative PAI features of human retinoblastoma and uveal melanoma have shown statistically significant difference in two tailed t-tests (p<0.05). Fully compatible with the concurrent procedures, PAI could be a potential tool complementary to other diagnostic modalities for characterizing intraocular tumors. PMID:28231293

  8. Separation of malignant and benign masses using image and segmentation features

    Kinnard, Lisa M.; Lo, Shih-Chung B.; Wang, Paul C.; Freedman, Matthew T.; Chouikha, Mohamed F.


    The purpose of this study is to investigate the efficacy of image features versus likelihood features of tumor boundaries for differentiating benign and malignant tumors and to compare the effectiveness of two neural networks in the classification study: (1) circular processing-based neural network and (2) conventional Multilayer Perceptron (MLP). The segmentation method used is an adaptive region growing technique coupled with a fuzzy shadow approach and maximum likelihood analyzer. Intensity, shape, texture, and likelihood features were calculated for the extracted Region of Interest (ROI). We performed these studies: experiment number 1 utilized image features used as inputs and the MLP for classification, experiment number 2 utilized image features used as inputs and the neural net with circular processing for classification, and experiment number 3 used likelihood values as inputs and the MLP for classification. The experiments were validated using an ROC methodology. We have tested these methods on 51 mammograms using a leave-one-case-out experiment (i.e., Jackknife procedure). The Az values for the four experiments were as follows: 0.66 in experiment number 1, 0.71 in experiment number 2, and 0.84 in experiment number 3.

  9. Featured Image: A New Look at Malin 1

    Kohler, Susanna


    Monochrome, inverted version of Malin 1. [Adapted from Galaz et al. 2015]The above image of Malin 1, the faintest and largest low-surface-brightness galaxy ever observed, was obtained with an instrument called Megacam on the 6.5m Magellan/Clay telescope. Gaspar Galaz (Pontifical Catholic University of Chile) and collaborators used Megacam to obtain deep optical observations of Malin 1. They then used novel noise-reduction and image-processing techniques to create this spectacular image of the spiral galaxy located roughly 1.2 billion light-years away. This new view of Malin 1 reveals details weve never before seen, including a stream within the disk that may have been caused by a past interaction between Malin 1 and another galaxy near it. Check outthe image to the rightfor a monochrome, inverted version thatmakes it a little easier to see some of Malin 1s features. To see the full original images and to learn more about what the images reveal about Malin 1, see the paper below.CitationGaspar Galaz et al 2015 ApJ 815 L29. doi:10.1088/2041-8205/815/2/L29

  10. Unusual acute encephalitis involving the thalamus: imaging features

    Kim, Sam Soo [Kangwon National University Hospital, Chuncheon (Korea, Republic of); Chang, Kee Hyun; Kim, Kyung Won; Han Moon Hee [Seoul National University College of Medicine, Seoul (Korea, Republic of); Park, Sung Ho; Nam, Hyun Woo [Seoul City Boramae Hospital, Seoul (Korea, Republic of); Choi, Kyu Ho [Kangnam St. Mary' s Hospital, Seoul (Korea, Republic of); Cho, Woo Ho [Sanggyo Paik Hospital, Seoul (Korea, Republic of)


    To describe the brain CT and MR imaging findings of unusual acute encephalitis involving the thalamus. We retrospectively reviewed the medical records and CT and/or MR imaging findings of six patients with acute encephalitis involving the thalamus. CT (n=6) and MR imaging (n=6) were performed during the acute and/or convalescent stage of the illness. Brain CT showed brain swelling (n=2), low attenuation of both thalami (n=1) or normal findings (n=3). Initial MR imaging indicated that in all patients the thalamus was involved either bilaterally (n=5) or unilaterally (n=1). Lesions were also present in the midbrain (n=5), medial temporal lobe (n=4), pons (n=3), both hippocampi (n=3) the insular cortex (n=2), medulla (n=2), lateral temporal lobe cortex (n=1), both cingulate gyri (n=1), both basal ganglia (n=1), and the left hemispheric cortex (n=1). These CT or MR imaging findings of acute encephalitis of unknown etiology were similar to a combination of those of Japanese encephalitis and herpes simplex encephalitis. In order to document the specific causative agents which lead to the appearance of these imaging features, further investigation is required.

  11. MR imaging features of focal liver lesions in Wilson disease.

    Dohan, Anthony; Vargas, Ottavia; Dautry, Raphael; Guerrache, Youcef; Woimant, France; Hamzi, Lounis; Boudiaf, Mourad; Poujois, Aurelia; Faraoun, Sid Ahmed; Soyer, Philippe


    Hepatic involvement in Wilson disease (WD) manifests as a diffuse chronic disease in the majority of patients. However, in a subset of patients focal liver lesions may develop, presenting with a wide range of imaging features. The majority of focal liver lesions in patients with WD are benign nodules, but there are reports that have described malignant liver tumors or dysplastic nodules in these patients. Because of the possibility of malignant transformation of liver nodules, major concerns have been raised with respect to the management and follow-up of patients with WD in whom focal liver lesions have been identified. The assessment of liver involvement in patients with WD is generally performed with ultrasonography. However, ultrasonography conveys limited specificity so that magnetic resonance (MR) imaging is often performed to improve lesion characterization. This review was performed to illustrate the spectrum of MR imaging features of focal liver lesions that develop in patients with WD. It is assumed that familiarity with the MR imaging presentation of focal liver lesions in WD may help clarify the actual nature of hepatic nodules in patients with this condition.

  12. Medical Image Fusion Based on Feature Extraction and Sparse Representation.

    Fei, Yin; Wei, Gao; Zongxi, Song


    As a novel multiscale geometric analysis tool, sparse representation has shown many advantages over the conventional image representation methods. However, the standard sparse representation does not take intrinsic structure and its time complexity into consideration. In this paper, a new fusion mechanism for multimodal medical images based on sparse representation and decision map is proposed to deal with these problems simultaneously. Three decision maps are designed including structure information map (SM) and energy information map (EM) as well as structure and energy map (SEM) to make the results reserve more energy and edge information. SM contains the local structure feature captured by the Laplacian of a Gaussian (LOG) and EM contains the energy and energy distribution feature detected by the mean square deviation. The decision map is added to the normal sparse representation based method to improve the speed of the algorithm. Proposed approach also improves the quality of the fused results by enhancing the contrast and reserving more structure and energy information from the source images. The experiment results of 36 groups of CT/MR, MR-T1/MR-T2, and CT/PET images demonstrate that the method based on SR and SEM outperforms five state-of-the-art methods.


    S. Praveenkumar


    Full Text Available Currently, medical image processing draws intense interests of scien- tists and physicians to aid in clinical diagnosis. The retinal Fundus image is widely used in the diagnosis and treatment of various eye diseases such as Diabetic Retinopathy, glaucoma etc. If these diseases are detected and treated early, many of the visual losses can be pre- vented. This paper presents the methods to detect main features of Fundus images such as optic disk, fovea, exudates and blood vessels. To determine the optic Disk and its centre we find the brightest part of the Fundus. The candidate region of fovea is defined an area circle. The detection of fovea is done by using its spatial relationship with optic disk. Exudates are found using their high grey level variation and their contours are determined by means of morphological recon- struction techniques. The blood vessels are highlighted using bottom hat transform and morphological dilation after edge detection. All the enhanced features are then combined in the Fundus image for the detection of abnormalities in eye.

  14. Multispectral image feature fusion for detecting land mines

    Clark, G.A.; Fields, D.J.; Sherwood, R.J. [Lawrence Livermore National Lab., CA (United States)] [and others


    Our system fuses information contained in registered images from multiple sensors to reduce the effect of clutter and improve the the ability to detect surface and buried land mines. The sensor suite currently consists if a camera that acquires images in sixible wavelength bands, du, dual-band infrared (5 micron and 10 micron) and ground penetrating radar. Past research has shown that it is extremely difficult to distinguish land mines from background clutter in images obtained from a single sensor. It is hypothesized, however, that information fused from a suite of various sensors is likely to provide better detection reliability, because the suite of sensors detects a variety of physical properties that are more separate in feature space. The materials surrounding the mines can include natural materials (soil, rocks, foliage, water, holes made by animals and natural processes, etc.) and some artifacts.

  15. Pro duct Image Classification Based on Fusion Features

    YANG Xiao-hui; LIU Jing-jing; YANG Li-jun


    Two key challenges raised by a product images classification system are classi-fication precision and classification time. In some categories, classification precision of the latest techniques, in the product images classification system, is still low. In this paper, we propose a local texture descriptor termed fan refined local binary pattern, which captures more detailed information by integrating the spatial distribution into the local binary pattern feature. We compare our approach with different methods on a subset of product images on Amazon/eBay and parts of PI100 and experimental results have demonstrated that our proposed approach is superior to the current existing methods. The highest classification precision is increased by 21%and the average classification time is reduced by 2/3.

  16. Image Watermarking Using Visual Perception Model and Statistical Features



    Full Text Available This paper presents an effective method for the image watermarking using visual perception model based on statistical features in the low frequency domain. In the image watermarking community watermark resistance to geometric attacks is an important issue. Most countermeasures proposed in the literature usually focus on the problem of global affine transforms such as rotation, scaling and translation (RST, but few are resistant to challenging cropping and random bending attacks (RBAs. Normally in the case of watermarking there may be an occurrence of distortion in the form of artifacts. A visual perception model is proposed to quantify the localized tolerance to noise for arbitrary imagery which achieves the reduction of artifacts. As a result, the watermarking system provides a satisfactory performance for those content-preserving geometric deformations and image processing operations, including JPEG ompression, low pass filtering, cropping and RBAs.

  17. Image Relaxation Matching Based on Feature Points for DSM Generation

    ZHENG Shunyi; ZHANG Zuxun; ZHANG Jianqing


    In photogrammetry and remote sensing, image matching is a basic and crucial process for automatic DEM generation. In this paper we presented a image relaxation matching method based on feature points. This method can be considered as an extention of regular grid point based matching. It avoids the shortcome of grid point based matching. For example, with this method, we can avoid low or even no texture area where errors frequently appear in cross correlaton matching. In the mean while, it makes full use of some mature techniques such as probability relaxation, image pyramid and the like which have already been successfully used in grid point matching process. Application of the technique to DEM generaton in different regions proved that it is more reasonable and reliable.

  18. CFA-aware features for steganalysis of color images

    Goljan, Miroslav; Fridrich, Jessica


    Color interpolation is a form of upsampling, which introduces constraints on the relationship between neighboring pixels in a color image. These constraints can be utilized to substantially boost the accuracy of steganography detectors. In this paper, we introduce a rich model formed by 3D co-occurrences of color noise residuals split according to the structure of the Bayer color filter array to further improve detection. Some color interpolation algorithms, AHD and PPG, impose pixel constraints so tight that extremely accurate detection becomes possible with merely eight features eliminating the need for model richification. We carry out experiments on non-adaptive LSB matching and the content-adaptive algorithm WOW on five different color interpolation algorithms. In contrast to grayscale images, in color images that exhibit traces of color interpolation the security of WOW is significantly lower and, depending on the interpolation algorithm, may even be lower than non-adaptive LSB matching.

  19. Solitary fibrous tumors of the central nervous system: clinical features and imaging findings in 22 patients.

    Wang, Xiao-Qiang; Zhou, Qing; Li, Shi-Ting; Liao, Chen-Long; Zhang, Hua; Zhang, Bi-Yun


    Solitary fibrous tumor (SFT) is a rare mesenchymal neoplasm originating in the central nervous system (CNS), with imaging features currently not well known. The purposes were to describe and characterize clinical features and imaging findings of CNS SFT. We retrospectively reviewed computed tomographic (CT; n = 10) and magnetic resonance (MR) images (n = 18) of 22 patients with SFT (13 males and 9 females; mean, 47.6 years) with associated clinical records. Each lesion was found as a solitary, well-defined mass, ranging in size from 12 to 70 mm (mean, 38 mm). The tumor shape was roundlike in 16 cases (72.7%) and irregular in 6 cases (27.2%). The cerebellopontine angle zone was the most affected area (n = 6). On precontrast CT scans, 10 cases showed predominantly hyperattenuation (n = 9) and isoattenuation (n = 1). No lesion contained calcification, and 2 cases showed bone invasions. All 18 tumors examined by MR imaging showed homogeneous hypointensive (n = 5) or isointensive (n = 7) signal intensity and heterogeneous mixed isointense and hypointense signal intensity (n = 6) on T1-weighted images, whereas most tumors were predominantly isointense (n = 13) and hypointense (n = 4) to the cortex on T2-weighted images; on postcontrast CT and MR images, enhancement was marked homogeneous (n = 10) or heterogeneous (n = 12). Fourteen tumors had thickening of the meninges adjacent to the tumor. Although SFT is a rare neoplasm in the CNS, it should be considered in the differential diagnosis. The most affected area is the cerebellopontine angle zone. Solitary fibrous tumor tends to have some imaging features, such as high attenuation on CT, isointense to hypointense signal intensity on MR images, and marked enhancement.

  20. Injectable facial fillers: imaging features, complications, and diagnostic pitfalls at MRI and PET CT.

    Mundada, Pravin; Kohler, Romain; Boudabbous, Sana; Toutous Trellu, Laurence; Platon, Alexandra; Becker, Minerva


    Injectable fillers are widely used for facial rejuvenation, correction of disabling volumetric fat loss in HIV-associated facial lipoatrophy, Romberg disease, and post-traumatic facial disfiguring. The purpose of this article is to acquaint the reader with the anatomy of facial fat compartments, as well as with the properties and key imaging features of commonly used facial fillers, filler-related complications, interpretation pitfalls, and dermatologic conditions mimicking filler-related complications. The distribution of facial fillers is characteristic and depends on the anatomy of the superficial fat compartments. Silicone has signature MRI features, calcium hydroxyapatite has characteristic calcifications, whereas other injectable fillers have overlapping imaging features. Most fillers (hyaluronic acid, collagen, and polyalkylimide-polyacrylamide hydrogels) have signal intensity patterns compatible with high water content. On PET-CT, most fillers show physiologic high FDG uptake, which should not be confounded with pathology. Abscess, cellulitis, non-inflammatory nodules, and foreign body granulomas are the most common filler-related complications, and imaging can help in the differential diagnosis. Diffusion weighted imaging helps in detecting a malignant lesion masked by injected facial fillers. Awareness of imaging features of facial fillers and their complications helps to avoid misinterpretation of MRI, and PET-CT scans and facilitates therapeutic decisions in unclear clinical cases. • Facial fillers are common incidental findings on MRI and PET-CT scans. • They have a characteristic appearance and typical anatomic distribution • Although considered as safe, facial filler injections are associated with several complications • As they may mask malignancy, knowledge of typical imaging features is mandatory. • MRI is a problem-solving tool for unclear cases.

  1. Feature extraction for target identification and image classification of OMIS hyperspectral image

    DU Pei-jun; TAN Kun; SU Hong-jun


    In order to combine feature extraction operations with specific hyperspectrai remote sensing information processing objectives, two aspects of feature extraction were explored. Based on clustering and decision tree algorithm, spectral absorption index (SAI), continuum-removal and derivative spectral analysis were employed to discover characterized spectral features of dif-ferent targets, and decision trees for identifying a specific class and discriminating different classes were generated. By combining support vector machine (SVM) classifier with different feature extraction strategies including principal component analysis (PCA), minimum noise fraction (MNF), grouping PCA, and derivate spectral analysis, the performance of feature extraction approaches in classification was evaluated. The results show that feature extraction by PCA and derivate spectral analysis are effective to OMIS (operational modular imaging spectrometer) image classification using SVM, and SVM outperforms traditional SAM and MLC classifiers for OMIS data.

  2. Differentially-Enhanced Sideband Imaging via Radio-frequency Encoding

    Fard, A M; Jalali, B


    We present a microscope paradigm that performs differential interference imaging with high sensitivity via optical amplification and radio-frequency (RF) heterodyne detection. This method, termed differentially-enhanced sideband imaging via radio-frequency encoding (DESIRE), uniquely exploits frequency-to-space mapping technique to encode the image of an object onto the RF sidebands of an illumination beam. As a proof-of-concept, we show validation experiment by implementing radio frequency (f = 15 GHz) phase modulation in conjunction with spectrally-encoded laser scanning technique to acquire one-dimensional image of a barcode-like object using a commercial RF spectrum analyzer.

  3. Mucocele and fibroma: treatment and clinical features for differential diagnosis.

    Valério, Rodrigo Alexandre; de Queiroz, Alexandra Mussolino; Romualdo, Priscila Coutinho; Brentegani, Luiz Guilherme; de Paula-Silva, Francisco Wanderley Garcia


    Mucocele is a benign lesion occurring in the buccal mucosa as a result of the rupture of a salivary gland duct and consequent outpouring of mucin into soft tissue. It is usually caused by a local trauma, although in many cases the etiology is uncertain. Mucocele is more commonly found in children and young adults, and the most frequent site is the lower inner portion of the lips. Fibroma, on the other hand, is a benign tumor of fibrous connective tissue that can be considered a reactionary connective tissue hyperplasia in response to trauma and irritation. They usually present hard consistency, are nodular and asymptomatic, with a similar color to the mucosa, sessile base, smooth surface, located in the buccal mucosa along the line of occlusion, tongue and lip mucosa. Conventional treatment for both lesions is conservative surgical excision. Recurrence rate is low for fibroma and high for oral mucoceles. This report presents a series of cases of mucocele and fibroma treated by surgical excision or enucleation and the respective follow-up routine in the dental clinic and discusses the features to be considered in order to distinguish these lesions from each other.

  4. Imaging and differential diagnosis of pediatric spinal tuberculosis

    Xiao-ying Xing


    Conclusion: Pediatric spinal tuberculosis often occurs in the cervical and thoracic vertebrae with typical imaging findings. The cases with atypical manifestations should be differentiated from other diseases such as Langerhans cell histiocytosis and metastatic neoplasm.

  5. Shape Adaptive, Robust Iris Feature Extraction from Noisy Iris Images

    Ghodrati, Hamed; Dehghani, Mohammad Javad; Danyali, Habibolah


    In the current iris recognition systems, noise removing step is only used to detect noisy parts of the iris region and features extracted from there will be excluded in matching step. Whereas depending on the filter structure used in feature extraction, the noisy parts may influence relevant features. To the best of our knowledge, the effect of noise factors on feature extraction has not been considered in the previous works. This paper investigates the effect of shape adaptive wavelet transform and shape adaptive Gabor-wavelet for feature extraction on the iris recognition performance. In addition, an effective noise-removing approach is proposed in this paper. The contribution is to detect eyelashes and reflections by calculating appropriate thresholds by a procedure called statistical decision making. The eyelids are segmented by parabolic Hough transform in normalized iris image to decrease computational burden through omitting rotation term. The iris is localized by an accurate and fast algorithm based on coarse-to-fine strategy. The principle of mask code generation is to assign the noisy bits in an iris code in order to exclude them in matching step is presented in details. An experimental result shows that by using the shape adaptive Gabor-wavelet technique there is an improvement on the accuracy of recognition rate. PMID:24696801

  6. A feature-enriched completely blind image quality evaluator.

    Lin Zhang; Lei Zhang; Bovik, Alan C


    Existing blind image quality assessment (BIQA) methods are mostly opinion-aware. They learn regression models from training images with associated human subjective scores to predict the perceptual quality of test images. Such opinion-aware methods, however, require a large amount of training samples with associated human subjective scores and of a variety of distortion types. The BIQA models learned by opinion-aware methods often have weak generalization capability, hereby limiting their usability in practice. By comparison, opinion-unaware methods do not need human subjective scores for training, and thus have greater potential for good generalization capability. Unfortunately, thus far no opinion-unaware BIQA method has shown consistently better quality prediction accuracy than the opinion-aware methods. Here, we aim to develop an opinion-unaware BIQA method that can compete with, and perhaps outperform, the existing opinion-aware methods. By integrating the features of natural image statistics derived from multiple cues, we learn a multivariate Gaussian model of image patches from a collection of pristine natural images. Using the learned multivariate Gaussian model, a Bhattacharyya-like distance is used to measure the quality of each image patch, and then an overall quality score is obtained by average pooling. The proposed BIQA method does not need any distorted sample images nor subjective quality scores for training, yet extensive experiments demonstrate its superior quality-prediction performance to the state-of-the-art opinion-aware BIQA methods. The MATLAB source code of our algorithm is publicly available at

  7. The imaging features of neurologic complications of left atrial myxomas

    Liao, Wei-Hua; Ramkalawan, Divya; Liu, Jian-Ling; Shi, Wei [Department of Radiology, Xiangya Hospital, Central South University, Changsha 410008, Hunan (China); Zee, Chi-Shing [Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA 90033 (United States); Yang, Xiao-Su; Li, Guo-Liang; Li, Jing [Department of Neurology, Xiangya Hospital, Central South University, Changsha 410008, Hunan (China); Wang, Xiao-Yi, E-mail: [Department of Radiology, Xiangya Hospital, Central South University, Changsha 410008, Hunan (China)


    Background: Neurologic complications may be the first symptoms of atrial myxomas. Understanding the imaging features of neurologic complications of atrial myxomas can be helpful for the prompt diagnosis. Objective: To identify neuroimaging features for patients with neurologic complications attributed to atrial myxoma. Methods: We retrospectively reviewed the medical records of 103 patients with pathologically confirmed atrial myxoma at Xiangya Hospital from January 2009 to January 2014. The neuroimaging data for patients with neurologic complications were analyzed. Results: Eight patients with atrial myxomas (7.77%) presented with neurologic manifestations, which constituted the initial symptoms for seven patients (87.5%). Neuroimaging showed five cases of cerebral infarctions and three cases of aneurysms. The main patterns of the infarctions were multiplicity (100.0%) and involvement of the middle cerebral artery territory (80.0%). The aneurysms were fusiform in shape, multiple in number (100.0%) and located in the distal middle cerebral artery (100.0%). More specifically, high-density in the vicinity of the aneurysms was observed on CT for two patients (66.7%), and homogenous enhancement surrounding the aneurysms was detected in the enhanced imaging for two patients (66.7%). Conclusion: Neurologic complications secondary to atrial myxoma consist of cerebral infarctions and aneurysms, which show certain characteristic features in neuroimaging. Echocardiography should be performed in patients with multiple cerebral infarctions, and multiple aneurysms, especially when aneurysms are distal in location. More importantly, greater attention should be paid to the imaging changes surrounding the aneurysms when myxomatous aneurysms are suspected and these are going to be the relevant features in our article.

  8. MRI and PET Image Fusion Using Fuzzy Logic and Image Local Features

    Umer Javed


    to maximally combine useful information present in MRI and PET images. Image local features are extracted and combined with fuzzy logic to compute weights for each pixel. Simulation results show that the proposed scheme produces significantly better results compared to state-of-art schemes.

  9. Diagnostic features of lung metastases differentiated thyroid cancer

    T. M. Geliashvili


    Full Text Available Background. The worldwide increasing incidence of thyroid cancer (TC is mainly due to a rise in its major form of differentiated TC (DTC: papillary. Most patients with DTC have a good prognosis; 10-year survival overall rates are as high as 85 %, but not greater than 40 % in a group of patients with distant metastases. At the same time, the lung is the most frequent target for distant metastases, accounting for 70 % of all sites.Objective: to estimate and compare the capabilities of different diagnostic techniques to detect lung metastases of DTC. Materials and methods. The results of diagnosing lung metastases were retrospectively analyzed in 36 patients (33 women and 3 men; mean age 53 years with DTC (29 patients with papillary TC and 7 with follicular TC treated at the department of radiotherapy with systemic therapy, Chelyabinsk Regional Clinical Oncology Center from 2011 to 2014.Results. Chest X-ray could reveal pulmonary metastases in 13 (36 % patients; lung pathology foci were absent in 23 (64 % patients. 131I whole-body scintigraphy (WBS proved to be of informative value in 24 (66.7 % patients, it displayed no increased accumulation of the radiopharmaceutical in the lung of 12 (33.3 % cases. Multislice spiral computed tomography (MSCT of the chest was carried out in 22 (61 % patients; out of them 21 (95.5 % were found to have 1.4-to-20-mm lung cancer foci. 18Fluorodeoxyglucose (18FDG positron emission tomography / computed tomography (PET / CT was performed in 18 (50 % patients, which showed 3–26-mm lung pathology foci in all the patents; out of them 16 (88.9 % were detected to have metastases owing to the CT component of this method. Thus, the highest sensitivity was exhibited by MSCT (95.5 %, 18FDG PET / CT (100 % due to its CT component, and 131I WBS (66.7 %.Conclusion. When lung metastases of DTC are suspected, 1 chest X-ray should be used as a screening test; 2 131I WBS should be performed in all patients; 3 MSCT of the chest is

  10. The ultrasonographic features of endometriomas: morphologic analysis and differential diagnosis

    Kim, Mi Sung; Park, Chan Sup; Song, Soon Young; Lee, Eun Ja; Park, No Hyuck [College of Medicine, Kwandong Univ., Koyang (Korea, Republic of); Park, Cheol Min [College of Medicine, Korea Univ., Seoul (Korea, Republic of); Kim, Bo Hyun; Kim, Chan Kyo [College of Medicine, Sungkyunkwan Univ., Seoul (Korea, Republic of)


    septation, wall nodularity, focal echogenic wall foci, and a solid area, all of which were also apparent in group I. The US findings of endometriomas vary: the most common is homogeneous fine internal echoes (79%), found in 85% of unilocular or multiseptated cysts. Their appearance may also be atypical, however: namely solid and cystic or mixed type, with diverse internal echogenicity, and such masses should be differentiated from other adnexal masses such as cystic neoplasm, teratoma, hemorrhagic cyst, functional cyst and ovarian cancer.

  11. Appendiceal mucocele: clinical and imaging features of 14 cases.

    Malya, F Umit; Hasbahceci, M; Serter, A; Cipe, G; Karatepe, O; Kocakoc, E; Muslumanoglu, M


    Appendiceal mucocele as a cystic dilatation filled with mucinous material is a very rare disease of the appendix vermiformis. Its preoperative diagnosis is still acking behind common use of imaging techniques. Retrospective analysis of the patients with a pathological diagnosis of appendiceal mucocele with regard to clinical and imaging features. The study group included 14 patients with a mean age of 51 years (range from 17 to 82 years). Predominant symptoms were pain and feeling of fullness in the right iliac fossa in 9(64%) and 5 (36%) patients, respectively. For imaging purposes, use of computed tomography resulted in preoperative diagnosis of appendiceal mucocele in half of the patients(50%). 93% of the cases underwent appendectomy, and righth emicolectomy was performed in one patient (7%). Mucocele and cystadenoma were detected in 11 (79%) and 3 (21%)patients, respectively. Presence of acute appendicitis and coloncarcinoma were confirmed afterwards histologically in 4 (29%)and one (7%) patients, respectively. Despite the common use of imaging studies,preoperative diagnosis of appendiceal mucocele is still not possible in most of the cases. During surgical treatment,which is tailored according to imaging and intraoperative findings, precautionary measures to avoid intraperitoneal rupture and dissemination should be taken. Celsius.

  12. Lipofibromatosis: magnetic resonance imaging features and pathological correlation in three cases

    Vogel, Daniela; Righi, Alberto; Kreshak, Jennifer; Dei Tos, Angelo Paolo [Istituto Ortopedico Rizzoli, Bologna (Italy); Merlino, Biagio [Universita Cattolica del Sacro Cuore Policlinico ' ' A. Gemelli' ' , Dipartimento di Scienze Radiologiche, Roma (Italy); Brunocilla, Eugenio [U.O. di UROLOGIA, Dipartimento di Medicina Specialistica, Diagnostica e Sperimentale, Bologna (Italy); Vanel, Daniel [Istituto Ortopedico Rizzoli, Anatomia Patologica, Bologna (Italy)


    Lipofibromatosis is a rare, benign, but infiltrative, soft tissue tumor seen in children. We present three cases of lipofibromatosis, each with different magnetic resonance imaging features and correlate this with the histological findings. The patients comprised two males and one female who presented in infancy; at birth, 5 months, and 7 months of age. Clinically, the masses were painless and slow-growing. The masses ranged in size from 2 to 6 cm and involved the distal extremities in two cases (one foot, one wrist) and the trunk. Magnetic resonance imaging showed lipomatous lesions with varying amounts of adipose and solid components in each case. There were no capsules at the periphery of the lesions. One case showed a fat-predominant lesion, another an equal mixture of fat and solid tissue, and the third was predominantly solid. This was reflected in the histology, which showed corresponding features. Radiological and histopathological differential diagnoses are reviewed. (orig.)

  13. A novel point cloud registration using 2D image features

    Lin, Chien-Chou; Tai, Yen-Chou; Lee, Jhong-Jin; Chen, Yong-Sheng


    Since a 3D scanner only captures a scene of a 3D object at a time, a 3D registration for multi-scene is the key issue of 3D modeling. This paper presents a novel and an efficient 3D registration method based on 2D local feature matching. The proposed method transforms the point clouds into 2D bearing angle images and then uses the 2D feature based matching method, SURF, to find matching pixel pairs between two images. The corresponding points of 3D point clouds can be obtained by those pixel pairs. Since the corresponding pairs are sorted by their distance between matching features, only the top half of the corresponding pairs are used to find the optimal rotation matrix by the least squares approximation. In this paper, the optimal rotation matrix is derived by orthogonal Procrustes method (SVD-based approach). Therefore, the 3D model of an object can be reconstructed by aligning those point clouds with the optimal transformation matrix. Experimental results show that the accuracy of the proposed method is close to the ICP, but the computation cost is reduced significantly. The performance is six times faster than the generalized-ICP algorithm. Furthermore, while the ICP requires high alignment similarity of two scenes, the proposed method is robust to a larger difference of viewing angle.

  14. Feature detection on 3D images of dental imprints

    Mokhtari, Marielle; Laurendeau, Denis


    A computer vision approach for the extraction of feature points on 3D images of dental imprints is presented. The position of feature points are needed for the measurement of a set of parameters for automatic diagnosis of malocclusion problems in orthodontics. The system for the acquisition of the 3D profile of the imprint, the procedure for the detection of the interstices between teeth, and the approach for the identification of the type of tooth are described, as well as the algorithm for the reconstruction of the surface of each type of tooth. A new approach for the detection of feature points, called the watershed algorithm, is described in detail. The algorithm is a two-stage procedure which tracks the position of local minima at four different scales and produces a final map of the position of the minima. Experimental results of the application of the watershed algorithm on actual 3D images of dental imprints are presented for molars, premolars and canines. The segmentation approach for the analysis of the shape of incisors is also described in detail.

  15. Sparse coding based feature representation method for remote sensing images

    Oguslu, Ender

    In this dissertation, we study sparse coding based feature representation method for the classification of multispectral and hyperspectral images (HSI). The existing feature representation systems based on the sparse signal model are computationally expensive, requiring to solve a convex optimization problem to learn a dictionary. A sparse coding feature representation framework for the classification of HSI is presented that alleviates the complexity of sparse coding through sub-band construction, dictionary learning, and encoding steps. In the framework, we construct the dictionary based upon the extracted sub-bands from the spectral representation of a pixel. In the encoding step, we utilize a soft threshold function to obtain sparse feature representations for HSI. Experimental results showed that a randomly selected dictionary could be as effective as a dictionary learned from optimization. The new representation usually has a very high dimensionality requiring a lot of computational resources. In addition, the spatial information of the HSI data has not been included in the representation. Thus, we modify the framework by incorporating the spatial information of the HSI pixels and reducing the dimension of the new sparse representations. The enhanced model, called sparse coding based dense feature representation (SC-DFR), is integrated with a linear support vector machine (SVM) and a composite kernels SVM (CKSVM) classifiers to discriminate different types of land cover. We evaluated the proposed algorithm on three well known HSI datasets and compared our method to four recently developed classification methods: SVM, CKSVM, simultaneous orthogonal matching pursuit (SOMP) and image fusion and recursive filtering (IFRF). The results from the experiments showed that the proposed method can achieve better overall and average classification accuracies with a much more compact representation leading to more efficient sparse models for HSI classification. To further

  16. Evaluation of image features and classification methods for Barrett's cancer detection using VLE imaging

    Klomp, Sander; van der Sommen, Fons; Swager, Anne-Fré; Zinger, Svitlana; Schoon, Erik J.; Curvers, Wouter L.; Bergman, Jacques J.; de With, Peter H. N.


    Volumetric Laser Endomicroscopy (VLE) is a promising technique for the detection of early neoplasia in Barrett's Esophagus (BE). VLE generates hundreds of high resolution, grayscale, cross-sectional images of the esophagus. However, at present, classifying these images is a time consuming and cumbersome effort performed by an expert using a clinical prediction model. This paper explores the feasibility of using computer vision techniques to accurately predict the presence of dysplastic tissue in VLE BE images. Our contribution is threefold. First, a benchmarking is performed for widely applied machine learning techniques and feature extraction methods. Second, three new features based on the clinical detection model are proposed, having superior classification accuracy and speed, compared to earlier work. Third, we evaluate automated parameter tuning by applying simple grid search and feature selection methods. The results are evaluated on a clinically validated dataset of 30 dysplastic and 30 non-dysplastic VLE images. Optimal classification accuracy is obtained by applying a support vector machine and using our modified Haralick features and optimal image cropping, obtaining an area under the receiver operating characteristic of 0.95 compared to the clinical prediction model at 0.81. Optimal execution time is achieved using a proposed mean and median feature, which is extracted at least factor 2.5 faster than alternative features with comparable performance.

  17. Collaborative Tracking of Image Features Based on Projective Invariance

    Jiang, Jinwei

    -mode sensors for improving the flexibility and robustness of the system. From the experimental results during three field tests for the LASOIS system, we observed that most of the errors in the image processing algorithm are caused by the incorrect feature tracking. This dissertation addresses the feature tracking problem in image sequences acquired from cameras. Despite many alternatives to feature tracking problem, iterative least squares solution solving the optical flow equation has been the most popular approach used by many in the field. This dissertation attempts to leverage the former efforts to enhance feature tracking methods by introducing a view geometric constraint to the tracking problem, which provides collaboration among features. In contrast to alternative geometry based methods, the proposed approach provides an online solution to optical flow estimation in a collaborative fashion by exploiting Horn and Schunck flow estimation regularized by view geometric constraints. Proposed collaborative tracker estimates the motion of a feature based on the geometry of the scene and how the other features are moving. Alternative to this approach, a new closed form solution to tracking that combines the image appearance with the view geometry is also introduced. We particularly use invariants in the projective coordinates and conjecture that the traditional appearance solution can be significantly improved using view geometry. The geometric constraint is introduced by defining a new optical flow equation which exploits the scene geometry from a set drawn from tracked features. At the end of each tracking loop the quality of the tracked features is judged using both appearance similarity and geometric consistency. Our experiments demonstrate robust tracking performance even when the features are occluded or they undergo appearance changes due to projective deformation of the template. The proposed collaborative tracking method is also tested in the visual navigation

  18. Multimodal Ultrawide-Field Imaging Features in Waardenburg Syndrome.

    Choudhry, Netan; Rao, Rajesh C


    A 45-year-old woman was referred for bilateral irregular fundus pigmentation. Dilated fundus examination revealed irregular hypopigmentation posterior to the equator in both eyes, confirmed by fundus autofluorescence. A thickened choroid was seen on enhanced-depth imaging spectral-domain optical coherence tomography (EDI SD-OCT). Systemic evaluation revealed sensorineural deafness, telecanthus, and a white forelock. Further investigation revealed a first-degree relative with Waardenburg syndrome. Waardenburg syndrome is characterized by a group of features including telecanthus, a broad nasal root, synophrys of the eyebrows, piedbaldism, heterochromia irides, and deafness. Choroidal hypopigmentation is a unique feature that can be visualized with ultrawide-field fundus autofluorescence. The choroid may also be thickened and its thickness measured with EDI SD-OCT.

  19. Detecting curvatures in digital images using filters derived from differential geometry

    Toro Giraldo, Juanita


    Detection of curvature in digital images is an important theoretical and practical problem in image processing. Many important features in an image are associated with curvature and the detection of such features is reduced to detection and characterization of curvatures. Differential geometry studies many kinds of curvature operators and from these curvature operators is possible to derive powerful filters for image processing which are able to detect curvature in digital images and videos. The curvature operators are formulated in terms of partial differential operators which can be applied to images via convolution with generalized kernels derived from the the Korteweg- de Vries soliton . We present an algorithm for detection of curvature in digital images which is implemented using the Maple package ImageTools. Some experiments were performed and the results were very good. In a future research will be interesting to compare the results using the Korteweg-de Vries soliton with the results obtained using Airy derivatives. It is claimed that the resulting curvature detectors could be incorporated in standard programs for image processing.

  20. Imaging guided differentiation of parotid tumors; Bildgebende Differenzierung von Parotistumoren

    Kloth, C.; Horger, M.; Haap, M.; Ioanoviciu, S.D.; Boesmueller, H.


    Imaging guided differentiation of parotid tumors is helping diagnosis and therapy decision making. It is necessary to consider seldom tumor forms and their characteristic appearance. Modern techniques as diffusion supported NMR imaging sequences and correlated contrast agent kinetics may be helpful besides computer tomography and PET techniques.

  1. MR imaging features and staging of neuroendocrine carcinomas of the uterine cervix with pathological correlations

    Duan, Xiaohui; Zhang, Xiang; Hu, Huijun; Li, Guozhao; Wang, Dongye; Zhang, Fang; Shen, Jun [Sun Yat-Sen University, Department of Radiology, Sun Yat-Sen Memorial Hospital, Guangzhou (China); Ban, Xiaohua [Sun Yat-Sen University, Medical Imaging and Minimally Invasive Interventional Center and State Key Laboratory of Oncology in Southern China, Cancer Center, Guangzhou, Guangdong (China); Wang, Charles Qian [Sun Yat-Sen University, Department of Radiology, Sun Yat-Sen Memorial Hospital, Guangzhou (China); University of New South Wales, JMO, Westmead Hospital, Sydney (Australia)


    To determine MR imaging features and staging accuracy of neuroendocrine carcinomas (NECs) of the uterine cervix with pathological correlations. Twenty-six patients with histologically proven NECs, 60 patients with squamous cell carcinomas (SCCs), and 30 patients with adenocarcinomas of the uterine cervix were included. The clinical data, pathological findings, and MRI findings were reviewed retrospectively. MRI features of cervical NECs, SCCs, and adenocarcinomas were compared, and MRI staging of cervical NECs was compared with the pathological staging. Cervical NECs showed a higher tendency toward a homogeneous signal intensity on T2-weighted imaging and a homogeneous enhancement pattern, as well as a lower ADC value of tumour and a higher incidence of lymphadenopathy, compared with SCCs and adenocarcinomas (P < 0.05). An ADC value cutoff of 0.90 x 10{sup -3} mm{sup 2}/s was robust for differentiation between cervical NECs and other cervical cancers, with a sensitivity of 63.3 % and a specificity of 95 %. In 21 patients who underwent radical hysterectomy and lymphadenectomy, the overall accuracy of tumour staging by MR imaging was 85.7 % with reference to pathology staging. Homogeneous lesion texture and low ADC value are likely suggestive features of cervical NECs and MR imaging is reliable for the staging of cervical NECs. (orig.)

  2. Breast cancer mitosis detection in histopathological images with spatial feature extraction

    Albayrak, Abdülkadir; Bilgin, Gökhan


    In this work, cellular mitosis detection in histopathological images has been investigated. Mitosis detection is very expensive and time consuming process. Development of digital imaging in pathology has enabled reasonable and effective solution to this problem. Segmentation of digital images provides easier analysis of cell structures in histopathological data. To differentiate normal and mitotic cells in histopathological images, feature extraction step is very crucial step for the system accuracy. A mitotic cell has more distinctive textural dissimilarities than the other normal cells. Hence, it is important to incorporate spatial information in feature extraction or in post-processing steps. As a main part of this study, Haralick texture descriptor has been proposed with different spatial window sizes in RGB and La*b* color spaces. So, spatial dependencies of normal and mitotic cellular pixels can be evaluated within different pixel neighborhoods. Extracted features are compared with various sample sizes by Support Vector Machines using k-fold cross validation method. According to the represented results, it has been shown that separation accuracy on mitotic and non-mitotic cellular pixels gets better with the increasing size of spatial window.

  3. Imaging features of constrictive pericarditis: beyond pericardial thickening

    Napolitano, G.; Pressacco, J.; Paquet, E. [Dept. of Radiology, Montreal Heart Inst., Montreal, Quebec (Canada)], E-mail:


    Constrictive pericarditis is caused by adhesions between the visceral and parietal layers of the pericardium and progressive pericardial fibrosis that restricts diastolic filling of the heart. Later on, the thickened pericardium may calcify. Despite a better understanding of the pathophysiologic basis of the imaging findings in constrictive pericarditis and the recent advent of magnetic resonance imaging (MRI) technology, which has dramatically improved the visualization of the pericardium, the diagnosis of constrictive pericarditis remains a challenge in many cases. In patients with clinical suspicion of underlying constrictive pericarditis, the most important radiologic diagnostic feature is abnormal pericardial thickening, which can be shown readily by computed tomography (CT) and especially by MRI, and is highly suggestive of constrictive pericarditis. Nevertheless, a thickened pericardium does not always indicate constrictive pericarditis. Furthermore, constrictive pericarditis can occur without pericardial thickening. (author)

  4. Fourier domain image fusion for differential X-ray phase-contrast breast imaging.

    Coello, Eduardo; Sperl, Jonathan I; Bequé, Dirk; Benz, Tobias; Scherer, Kai; Herzen, Julia; Sztrókay-Gaul, Anikó; Hellerhoff, Karin; Pfeiffer, Franz; Cozzini, Cristina; Grandl, Susanne


    X-Ray Phase-Contrast (XPC) imaging is a novel technology with a great potential for applications in clinical practice, with breast imaging being of special interest. This work introduces an intuitive methodology to combine and visualize relevant diagnostic features, present in the X-ray attenuation, phase shift and scattering information retrieved in XPC imaging, using a Fourier domain fusion algorithm. The method allows to present complementary information from the three acquired signals in one single image, minimizing the noise component and maintaining visual similarity to a conventional X-ray image, but with noticeable enhancement in diagnostic features, details and resolution. Radiologists experienced in mammography applied the image fusion method to XPC measurements of mastectomy samples and evaluated the feature content of each input and the fused image. This assessment validated that the combination of all the relevant diagnostic features, contained in the XPC images, was present in the fused image as well.

  5. Osteosarcoma of the jaws: demographic and CT imaging features

    Wang, S; Shi, H; Yu, Q


    Objective The aim of this study was to evaluate the patient demographic and CT imaging findings of primary osteosarcoma of the jaws. Methods 88 primary osteosarcomas of the jaws histopathologically diagnosed during 1997–2007 were reviewed. 21 cases of CT images were reviewed. Results Of 88 patients, 51 (58%) had tumours in the mandible and 37 (42%) in the maxilla. The mean age was 37.8 years (range 9–80 years). The male-to-female ratio was 1.32:1. The mean age of patients with mandibular lesions was 41.04 years and in those with maxillary lesions it was 33.3 years. CT imaging findings were available in 21 patients. In the maxilla (n = 9), all tumours (100%) arose from the alveolar ridge. In the mandible (n = 12), most tumours (9 cases, 75%), arose from the ramus and/or condyle. All except two lesions had the epicentrum within the medullary cavity of the involved bone. The presence of periosteal reaction was demonstrated in 13 cases (62%). Soft-tissue extension was present in 18 lesions (86%), with calcification identified in 13 (72%). Conclusions This study provides age, sex distribution, location and CT imaging features of primary osteosarcoma of the jaws. PMID:22074870

  6. Feature identification for image-guided transcatheter aortic valve implantation

    Lang, Pencilla; Rajchl, Martin; McLeod, A. Jonathan; Chu, Michael W.; Peters, Terry M.


    Transcatheter aortic valve implantation (TAVI) is a less invasive alternative to open-heart surgery, and is critically dependent on imaging for accurate placement of the new valve. Augmented image-guidance for TAVI can be provided by registering together intra-operative transesophageal echo (TEE) ultrasound and a model derived from pre-operative CT. Automatic contour delineation on TEE images of the aortic root is required for real-time registration. This study develops an algorithm to automatically extract contours on simultaneous cross-plane short-axis and long-axis (XPlane) TEE views, and register these features to a 3D pre-operative model. A continuous max-flow approach is used to segment the aortic root, followed by analysis of curvature to select appropriate contours for use in registration. Results demonstrate a mean contour boundary distance error of 1.3 and 2.8mm for the short and long-axis views respectively, and a mean target registration error of 5.9mm. Real-time image guidance has the potential to increase accuracy and reduce complications in TAVI.

  7. High-Precision Image Aided Inertial Navigation with Known Features: Observability Analysis and Performance Evaluation

    Jiang, Weiping; Wang, Li; Niu, Xiaoji; Zhang, Quan; Zhang, Hui; Tang, Min; Hu, Xiangyun


    A high-precision image-aided inertial navigation system (INS) is proposed as an alternative to the carrier-phase-based differential Global Navigation Satellite Systems (CDGNSSs) when satellite-based navigation systems are unavailable. In this paper, the image/INS integrated algorithm is modeled by a tightly-coupled iterative extended Kalman filter (IEKF). Tightly-coupled integration ensures that the integrated system is reliable, even if few known feature points (i.e., less than three) are observed in the images. A new global observability analysis of this tightly-coupled integration is presented to guarantee that the system is observable under the necessary conditions. The analysis conclusions were verified by simulations and field tests. The field tests also indicate that high-precision position (centimeter-level) and attitude (half-degree-level)-integrated solutions can be achieved in a global reference. PMID:25330046

  8. High-Precision Image Aided Inertial Navigation with Known Features: Observability Analysis and Performance Evaluation

    Weiping Jiang


    Full Text Available A high-precision image-aided inertial navigation system (INS is proposed as an alternative to the carrier-phase-based differential Global Navigation Satellite Systems (CDGNSSs when satellite-based navigation systems are unavailable. In this paper, the image/INS integrated algorithm is modeled by a tightly-coupled iterative extended Kalman filter (IEKF. Tightly-coupled integration ensures that the integrated system is reliable, even if few known feature points (i.e., less than three are observed in the images. A new global observability analysis of this tightly-coupled integration is presented to guarantee that the system is observable under the necessary conditions. The analysis conclusions were verified by simulations and field tests. The field tests also indicate that high-precision position (centimeter-level and attitude (half-degree-level-integrated solutions can be achieved in a global reference.

  9. Analysis of muscle fatigue conditions using time-frequency images and GLCM features

    Karthick P.A.


    Full Text Available In this work, an attempt has been made to differentiate muscle non-fatigue and fatigue conditions using sEMG signals and texture representation of the time-frequency images. The sEMG signals are recorded from the biceps brachii muscle of 25 healthy adult volunteers during dynamic fatiguing contraction. The first and last curls of these signals are considered as the non-fatigue and fatigue zones, respectively. These signals are preprocessed and the time-frequency spectrum is computed using short time fourier transform (STFT. Gray-Level Co-occurrence Matrix (GLCM is extracted from low (15–45 Hz, medium (46–95 Hz and high (96–150 Hz frequency bands of the time-frequency images. Further, the features such as contrast, correlation, energy and homogeneity are calculated from the resultant matrices. The results show that the high frequency band based features are able to differentiate non-fatigue and fatigue conditions. The features such as correlation, contrast and homogeneity extracted at angles 0°, 45°, 90°, and 135° are found to be distinct with high statistical significance (p < 0.0001. Hence, this framework can be used for analysis of neuromuscular disorders.

  10. Image Retrieval based on combined features of image sub-blocks

    Ch.Kavitha,; Dr. B.Prabhakara Rao,; Dr. A. Govardhan3


    In this paper we propose a new and efficient technique to retrieve images based on sum of the values of Local Histogram and GLCM (Gray Level Co-occurrence Matrix) texture of image sub-blocks to enhance theretrieval performance. The image is divided into sub blocks of equal size. Then the color and texture features of each sub-block are computed. Most of the image retrieval techniques used Histograms for indexing. Histograms describe global intensity distribution. They are very easy to compute...

  11. Angioleiomyoma: magnetic resonance imaging features in ten cases

    Gupte, Cynthia; Butt, Sajid H.; Saifuddin, Asif [Royal National Orthopaedic Hospital, Department of Radiology, Middlesex (United Kingdom); Tirabosco, Roberto [Royal National Orthopaedic Hospital, Department of Histopathology, Middlesex (United Kingdom)


    Angioleiomyoma is a rare, benign smooth muscle tumour arising from the tunica media of small veins and arteries and can occur anywhere in the body. The histological appearances are well documented, but there are relatively few descriptions of the magnetic resonance imaging (MRI) findings. A retrospective study of the clinical presentation, MRI appearances and histological findings of ten angioleiomyomas presenting as extremity soft tissue masses. MRI typically demonstrated a well-defined, oval mass located superficial to the fascia with the commonest sites being the hand (three cases) and ankle/foot (five cases). The lesion was isointense to muscle on T1-weighted spin echo images with heterogeneous increased internal T2W/short tau inversion recovery (STIR) signal intensity, which commonly appeared as multiple linear or branching areas of hyperintensity. Enhancement after IV gadolinium ranged from diffuse to heterogeneous. In a single case, central fat signal intensity was seen, while a further case showed marked T2W/STIR hypointensity due to diffuse hyalinisation within the lesion. This is the largest reported MRI series of extremity musculoskeletal angioleiomyoma. Angioleiomyoma should be considered in the differential diagnosis of a superficial mass in the hand or foot, particularly when characteristic linear or branching hyperintensity is seen on T2W or STIR images. (orig.)

  12. Feature selection applied to ultrasound carotid images segmentation.

    Rosati, Samanta; Molinari, Filippo; Balestra, Gabriella


    The automated tracing of the carotid layers on ultrasound images is complicated by noise, different morphology and pathology of the carotid artery. In this study we benchmarked four methods for feature selection on a set of variables extracted from ultrasound carotid images. The main goal was to select those parameters containing the highest amount of information useful to classify the pixels in the carotid regions they belong to. Six different classes of pixels were identified: lumen, lumen-intima interface, intima-media complex, media-adventitia interface, adventitia and adventitia far boundary. The performances of QuickReduct Algorithm (QRA), Entropy-Based Algorithm (EBR), Improved QuickReduct Algorithm (IQRA) and Genetic Algorithm (GA) were compared using Artificial Neural Networks (ANNs). All methods returned subsets with a high dependency degree, even if the average classification accuracy was about 50%. Among all classes, the best results were obtained for the lumen. Overall, the four methods for feature selection assessed in this study return comparable results. Despite the need for accuracy improvement, this study could be useful to build a pre-classifier stage for the optimization of segmentation performance in ultrasound automated carotid segmentation.

  13. Texture features analysis for coastline extraction in remotely sensed images

    De Laurentiis, Raimondo; Dellepiane, Silvana G.; Bo, Giancarlo


    The accurate knowledge of the shoreline position is of fundamental importance in several applications such as cartography and ships positioning1. Moreover, the coastline could be seen as a relevant parameter for the monitoring of the coastal zone morphology, as it allows the retrieval of a much more precise digital elevation model of the entire coastal area. The study that has been carried out focuses on the development of a reliable technique for the detection of coastlines in remotely sensed images. An innovative approach which is based on the concepts of fuzzy connectivity and texture features extraction has been developed for the location of the shoreline. The system has been tested on several kind of images as SPOT, LANDSAT and the results obtained are good. Moreover, the algorithm has been tested on a sample of a SAR interferogram. The breakthrough consists in the fact that the coastline detection is seen as an important features in the framework of digital elevation model (DEM) retrieval. In particular, the coast could be seen as a boundary line all data beyond which (the ones representing the sea) are not significant. The processing for the digital elevation model could be refined, just considering the in-land data.

  14. Robust image analysis with sparse representation on quantized visual features.

    Bao, Bing-Kun; Zhu, Guangyu; Shen, Jialie; Yan, Shuicheng


    Recent techniques based on sparse representation (SR) have demonstrated promising performance in high-level visual recognition, exemplified by the highly accurate face recognition under occlusion and other sparse corruptions. Most research in this area has focused on classification algorithms using raw image pixels, and very few have been proposed to utilize the quantized visual features, such as the popular bag-of-words feature abstraction. In such cases, besides the inherent quantization errors, ambiguity associated with visual word assignment and misdetection of feature points, due to factors such as visual occlusions and noises, constitutes the major cause of dense corruptions of the quantized representation. The dense corruptions can jeopardize the decision process by distorting the patterns of the sparse reconstruction coefficients. In this paper, we aim to eliminate the corruptions and achieve robust image analysis with SR. Toward this goal, we introduce two transfer processes (ambiguity transfer and mis-detection transfer) to account for the two major sources of corruption as discussed. By reasonably assuming the rarity of the two kinds of distortion processes, we augment the original SR-based reconstruction objective with l(0) norm regularization on the transfer terms to encourage sparsity and, hence, discourage dense distortion/transfer. Computationally, we relax the nonconvex l(0) norm optimization into a convex l(1) norm optimization problem, and employ the accelerated proximal gradient method to optimize the convergence provable updating procedure. Extensive experiments on four benchmark datasets, Caltech-101, Caltech-256, Corel-5k, and CMU pose, illumination, and expression, manifest the necessity of removing the quantization corruptions and the various advantages of the proposed framework.

  15. Analysis and Reliability Performance Comparison of Different Facial Image Features

    J. Madhavan


    Full Text Available This study performs reliability analysis on the different facial features with weighted retrieval accuracy on increasing facial database images. There are many methods analyzed in the existing papers with constant facial databases mentioned in the literature review. There were not much work carried out to study the performance in terms of reliability and also how the method will perform on increasing the size of the database. In this study certain feature extraction methods were analyzed on the regular performance measure and also the performance measures are modified to fit the real time requirements by giving weight ages for the closer matches. In this study four facial feature extraction methods are performed, they are DWT with PCA, LWT with PCA, HMM with SVD and Gabor wavelet with HMM. Reliability of these methods are analyzed and reported. Among all these methods Gabor wavelet with HMM gives more reliability than other three methods performed. Experiments are carried out to evaluate the proposed approach on the Olivetti Research Laboratory (ORL face database.

  16. Schizencephaly: clinical and imaging features in 30 infantile cases.

    Denis, D; Chateil, J F; Brun, M; Brissaud, O; Lacombe, D; Fontan, D; Flurin, V; Pedespan, J


    Schizencephaly is an uncommon structural disorder of cerebral cortical development, characterized by congenital clefts spanning the cerebral hemispheres from the pial surface to the lateral ventricles and lined by cortical gray matter. Either an antenatal environmental incident or a genetic origin could be responsible for this lesion which occurs between the third and fourth month of gestation. We report the clinical and cranial imaging features of 30 children, of whom 15 had unilateral and 15 had bilateral lesions. Their ages at the time of the first presentation ranged from 1 month to 10 years. They were thoroughly studied from clinical, epileptical, imaging and electroencephalographic (EEG) viewpoints. Five patients were investigated by cranial computed tomography (CT), eight by cranial magnetic resonance (MR) imaging, and 17 by both methods. The clinical features consisted of mild hemiparesis in 17 cases (57%), 12/17 were related to a unilateral phenotype (80% of all unilateral forms) and 5/17 to a bilateral phenotype. A tetraparesis was present in nine cases, all of which were due to a bilateral cleft. Bilateral forms were significantly associated with tetraparesis, whereas unilateral forms were associated with hemiparesis. Mental retardation was observed in 17 cases (57%), and was observed significantly more often in bilateral clefts (80%). When both hemispheres are involved, an absence of reorganization of the brain function between the two hemispheres leads to severe mental deficits, in addition to the cerebral anomaly itself. Eleven patients had seizures (seven from unilateral and three from bilateral forms). The degree of malformation was not related to the severity of epilepsy. Migration disorders, such as dysplasia or heterotopia, were observed in 30% of cases and are also important etiopathogenetic factors. The septum pellucidum was absent in 13 cases (43%), with septo-optical dysplasia in two cases. Corpus callosum dysgenesis was noted in 30% of cases


    M. Mary Helta Daisy


    Full Text Available Image retrieval is a challenging and important research applications like digital libraries and medical image databases. Content-based image retrieval is useful in retrieving images from database based on the feature vector generated with the help of the image features. In this study, we present image retrieval based on the genetic algorithm. The shape feature and morphological based texture features are extracted images in the database and query image. Then generating chromosome based on the distance value obtained by the difference feature vector of images in the data base and the query image. In the selected chromosome the genetic operators like cross over and mutation are applied. After that the best chromosome selected and displays the most similar images to the query image. The retrieval performance of the method shows better retrieval result.

  18. Differentiating emotional responses to images and words

    Jensen, Camilla Birgitte Falk; Petersen, Michael Kai; Larsen, Jakob Eg

    series responses in a single subject based on only a few trials. Comparing our results against previous findings we identify multiple early and late ICA components that are similarly modulated by neutral, pleasant and unpleasant content in both images and words. Suggesting that we might be able to model......The emergence of low cost electroencephalography (EEG) wireless neuroheadsets may potentially turn smartphones into pocketable labs [1], and enable design of personalized interfaces that adapt the selection of media to our emotional responses when viewing images and reading text. However such EEG...... responses are characterized by only small voltage changes that have typically been found in group studies involving multiple trials and large numbers of participants. Hypothesizing that spatial filtering might enhance retrieval, we apply independent component analysis (ICA) to cluster scalp maps and time...

  19. Hepatic hemangiosarcoma: imaging findings and differential diagnosis

    Rademaker, J.; Galanski, M. [Department of Radiology I, Medical School Hannover (Germany); Widjaja, A. [Department of Gastroenterology and Hepatology, Medical School Hannover (Germany)


    Primary hepatic angiosarcoma is a rare mesenchymal tumor of the liver that usually presents with nonspecific symptoms in elderly men. We present four cases of hepatic hemangiosarcoma and discuss the imaging characteristics of this entity. Our series shows that this tumor is not uncommon in younger patients with no associated risk factors such as previous exposure to thorotrast or vinyl chloride. Our experiences on a limited number of patients suggests that the combined use of angiography and dual-phase helical CT provides a better identification of the tumor and its complications. Analysis of imaging studies in patients with hepatic hemangiosarcoma reveals hypervascular lesions. Common complications were portal vein thrombosis, Budd-Chiari syndrome, as well as arterio-venous or arterio-portal shunts. Due to the vascularity of the tumor, percutaneous liver biopsy is hazardous. (orig.)

  20. Featured Image: A Detailed Look at the Crab Nebula

    Kohler, Susanna


    Planning on watching fireworks tomorrow? Heres an astronomical firework to help you start the celebrations! A new study has stunningly detailed the Crab Nebula (click for a closer look), a nebula 6,500 light-years away thought to have been formedby a supernova explosion and the subsequent ultrarelativistic wind emitted by the pulsar at its heart. Led by Gloria Dubner (University of Buenos Aires), the authors of this study obtained new observations of the Crab Nebula from five different telescopes. They compiled these observations to compare the details of the nebulas structure across different wavelengths, which allowedthem to learnabout the sources of various features within the nebula. In the images above, thetop left shows the 3 GHz data from the Very Large Array (radio). Moving clockise, the radio data (shown in red) is composited with: infrared data from Spitzer Space Telescope, optical continuum from Hubble Space Telescope, 500-nm optical datafrom Hubble, and ultraviolet data from XMM-Newton. The final two images are of the nebula center, and they are composites of the radio imagewith X-ray data from Chandra and near-infrared data from Hubble. To read more about what Dubner and collaborators learned (and to see more spectacular images!), check out the paper below.CitationG. Dubner et al 2017 ApJ 840 82. doi:10.3847/1538-4357/aa6983

  1. Robust Colour Image Watermarking Scheme Based on Feature Points and Image Normalization in DCT Domain

    Ibrahim Alsonosi Nasir


    Full Text Available Geometric attacks can desynchronize the location of the watermark and hence cause incorrect watermark detection. This paper presents a robust c olour image watermarking scheme based on visually significant feature points and image norma lization technique. The feature points are used as synchronization marks between watermark emb edding and detection. The watermark is embedded into the non overlapped normalized circula r regions in the luminance component or the blue component of a color image. The embedding of the watermark is carried out by modifying the DCT coefficients values in selected b locks. The original unmarked image is not required for watermark extraction Experimental results show that the proposed scheme successfully makes the watermark perceptually invis ible as well as robust to common signal processing and geometric attacks.

  2. Multidetector Computed Tomography Features in Differentiating Exophytic Renal Angiomyolipoma from Retroperitoneal Liposarcoma

    Wang, Qiushi; Juan, Yu-Hsiang; Li, Yong; Xie, Jia-Jun; Liu, Hui; Huang, Hongfei; Liu, Zaiyi; Zheng, Junhui; Saboo, Ujwala S.; Saboo, Sachin S.; Liang, Changhong


    Abstract This study aims to evaluate the multidetector computed tomography (CT) imaging features in differentiating exophytic renal angiomyolipoma (AML) from retroperitoneal liposarcoma. We retrospectively enrolled 42 patients with confirmed exophytic renal AML (31 patients) or retroperitoneal liposarcoma (11 patients) during 8 years period to assess: renal parenchymal defect at site of tumor contact, supply from branches of renal artery, tumoral vessel extending through the renal parenchyma, dilated intratumoral vessels, hemorrhage, non–fat-containing intratumoral nodules with postcontrast enhancement, calcification, renal sinus enlargement, anterior displacement of kidneys, and other associated AML. Renal parenchymal defect, renal arterial blood supply, tumoral vessel through the renal parenchyma, dilated intratumoral vessels, intratumoral/perirenal hemorrhage, renal sinus enlargement, and associated AML were seen only or mainly in exophytic renal AML (all P value < 0.05); however, non–fat-attenuating enhancing intratumoral nodules, intratumoral calcification, and anterior displacement of the kidney were more common in liposarcoma (all P value < 0.05). AMLs reveal renal parenchymal defect at the site of tumor contact, supply from renal artery, tumoral vessel extending through the renal parenchyma, dilated intratumoral vessels, intratumoral and/or perirenal hemorrhage, renal sinus enlargement, and associated AML. Non–fat-attenuating enhancing intratumoral nodules, intratumoral calcifications, and anterior displacement of kidney were more commonly seen in liposarcoma. PMID:26376398

  3. Non-interferometer Phase-differential Imaging Method with a Single Telescope Installation

    Choi, Jaeho


    Non-interferometer phase-differential imaging method for direct imaging of the astronomical objects will be presented. The feasibility of non-interferometry method to retrieve the phase differential images of the astronomical objects is demonstrated in the laboratory experiments exploiting the two-dimensional Foucault knife-edge filtering method which is installed on a single telescope. The experiment setup is essentially analogous to the Schlieren imaging apparatus that can be taken images using an incoherent light source. The fractional derivation filtering by the two-dimensional knife-edge filter is developed in order to acquire the phase information of the object. The intensities of filtering images by the 2D knife-edge at several points along the optical axis of the telescope are substituted in the transport-intensity equation to obtain phase-differential images of the astronomical objects. Then the phase-differential images are obtained by two image intensities taken along the optical axis. In our experiment, a mono-directional scanning scheme of the 2DFK was exploited to reduce number of scan as well as increase the spatial resolution of images. An illuminated light out of a bundle of optical fibers as an artificial astronomical object is used our laboratory based experiment. The light from the each optical fibers in the fiber bundle that intensities have exiguously different or barely visible are represented the brightness of the astronomical objects. The experiment result, the phase contrast images, shows that barely identified object from an intensity based image has rendered almost equivalent contrast as the bright object. It represents that our proposed method can be recovered from phase difference of the object light that could not be identified from the intensity of objects brightness. The proposed method has a feature of render phase-differential images as well as compensates atmospheric turbulence with the setup mounting on a single-telescope. The

  4. SU-E-J-237: Image Feature Based DRR and Portal Image Registration

    Wang, X; Chang, J [NY Weill Cornell Medical Ctr, NY (United States)


    Purpose: Two-dimensional (2D) matching of the kV X-ray and digitally reconstructed radiography (DRR) images is an important setup technique for image-guided radiotherapy (IGRT). In our clinics, mutual information based methods are used for this purpose on commercial linear accelerators, but with often needs for manual corrections. This work proved the feasibility that feature based image transform can be used to register kV and DRR images. Methods: The scale invariant feature transform (SIFT) method was implemented to detect the matching image details (or key points) between the kV and DRR images. These key points represent high image intensity gradients, and thus the scale invariant features. Due to the poor image contrast from our kV image, direct application of the SIFT method yielded many detection errors. To assist the finding of key points, the center coordinates of the kV and DRR images were read from the DICOM header, and the two groups of key points with similar relative positions to their corresponding centers were paired up. Using these points, a rigid transform (with scaling, horizontal and vertical shifts) was estimated. We also artificially introduced vertical and horizontal shifts to test the accuracy of our registration method on anterior-posterior (AP) and lateral pelvic images. Results: The results provided a satisfactory overlay of the transformed kV onto the DRR image. The introduced vs. detected shifts were fit into a linear regression. In the AP image experiments, linear regression analysis showed a slope of 1.15 and 0.98 with an R2 of 0.89 and 0.99 for the horizontal and vertical shifts, respectively. The results are 1.2 and 1.3 with R2 of 0.72 and 0.82 for the lateral image shifts. Conclusion: This work provided an alternative technique for kV to DRR alignment. Further improvements in the estimation accuracy and image contrast tolerance are underway.

  5. Fourth-order partial differential equations for effective image denoising

    Seongjai Kim


    Full Text Available This article concerns mathematical image denoising methods incorporating fourth-order partial differential equations (PDEs. We introduce and analyze piecewise planarity conditions (PPCs with which unconstrained fourth-order variational models in continuum converge to a piecewise planar image. It has been observed that fourth-order variational models holding PPCs can restore better images than models without PPCs and second-order models. Numerical schemes are presented in detail and various examples in image denoising are provided to verify the claim.

  6. Susceptibility weighted imaging: differentiating between calcification and hemosiderin

    Barbosa, Jeam Haroldo Oliveira; Salmon, Carlos Ernesto Garrido, E-mail: [Universidade de Sao Paulo (FFCLRP/USP), Ribeirao Preto, SP (Brazil). Faculdade de Filosofia, Ciencias e Letras; Santos, Antonio Carlos [Universidade de Sao Paulo (FMRP/USP), Ribeirao Preto, SP (Brazil). Faculdade de Medicina


    Objective: to present a detailed explanation on the processing of magnetic susceptibility weighted imaging (SWI), demonstrating the effects of echo time and sensitive mask on the differentiation between calcification and hemosiderin. Materials and methods: computed tomography and magnetic resonance (magnitude and phase) images of six patients (age range 41-54 years; four men) were retrospectively selected. The SWI images processing was performed using the Matlab's own routine. Results: four out of the six patients showed calcifications at computed tomography images and their SWI images demonstrated hyperintense signal at the calcification regions. The other patients did not show any calcifications at computed tomography, and SWI revealed the presence of hemosiderin deposits with hypointense signal. Conclusion: the selection of echo time and of the mask may change all the information on SWI images, and compromise the diagnostic reliability. Amongst the possible masks, the authors highlight that the sigmoid mask allows for contrasting calcifications and hemosiderin on a single SWI image. (author)

  7. Efficient image compression scheme based on differential coding

    Zhu, Li; Wang, Guoyou; Liu, Ying


    Embedded zerotree (EZW) and Set Partitioning in Hierarchical Trees (SPIHT) coding, introduced by J.M. Shapiro and Amir Said, are very effective and being used in many fields widely. In this study, brief explanation of the principles of SPIHT was first provided, and then, some improvement of SPIHT algorithm according to experiments was introduced. 1) For redundancy among the coefficients in the wavelet region, we propose differential method to reduce it during coding. 2) Meanwhile, based on characteristic of the coefficients' distribution in subband, we adjust sorting pass and optimize differential coding, in order to reduce the redundancy coding in each subband. 3) The image coding result, calculated by certain threshold, shows that through differential coding, the rate of compression get higher, and the quality of reconstructed image have been get raised greatly, when bpp (bit per pixel)=0.5, PSNR (Peak Signal to Noise Ratio) of reconstructed image exceeds that of standard SPIHT by 0.2~0.4db.

  8. Differential diagnosis of breast cancer using quantitative, label-free and molecular vibrational imaging.

    Yang, Yaliang; Li, Fuhai; Gao, Liang; Wang, Zhiyong; Thrall, Michael J; Shen, Steven S; Wong, Kelvin K; Wong, Stephen T C


    We present a label-free, chemically-selective, quantitative imaging strategy to identify breast cancer and differentiate its subtypes using coherent anti-Stokes Raman scattering (CARS) microscopy. Human normal breast tissue, benign proliferative, as well as in situ and invasive carcinomas, were imaged ex vivo. Simply by visualizing cellular and tissue features appearing on CARS images, cancerous lesions can be readily separated from normal tissue and benign proliferative lesion. To further distinguish cancer subtypes, quantitative disease-related features, describing the geometry and distribution of cancer cell nuclei, were extracted and applied to a computerized classification system. The results show that in situ carcinoma was successfully distinguished from invasive carcinoma, while invasive ductal carcinoma (IDC) and invasive lobular carcinoma were also distinguished from each other. Furthermore, 80% of intermediate-grade IDC and 85% of high-grade IDC were correctly distinguished from each other. The proposed quantitative CARS imaging method has the potential to enable rapid diagnosis of breast cancer.

  9. Magnetic Resonance Imaging and DWI Features of Orbital Rhabdomyosarcoma

    Xuetao Mu; Hong Wang; Yueyue Li; Yuwen Hao; Chunnan Wu; Lin Ma


    Purpose:.To describe the magnetic resonance imaging (MRI) features of orbital rhabdomyosarcoma (RMS). Methods:.Thirty-nine patients with histopathologically con-firmed orbital RMS were retrospectively reviewed. All patients underwent orbital conventional MRI, including axial,.sagittal, and coronal T1-weighted,.T2-weighted,.and postcontrast T1-weighted sequences. The location, shape, margin, and MRI signal of the 39 lesions were reviewed..DWI in 15 patients and susceptibility weighted imaging (SWI) in 2 patients were also analyzed. Results:.Orbital MRI was available in 39 patients and re-vealed a soft tissue mass in the orbital region in all cases..Of the 39 patients,.the primary tumor sites were limited to the orbital proper in 31 cases, while 28 cases had extraocular muscle invasion and 8 cases had extraorbital invasion..All le-sions were unilateral..Thirty-three cases were well-defined soft tissue masses and 6 cases appeared as less well-defined soft-tissue masses..Thirty-four cases showed homogeneous isoin-tense or slightly hypointense signals on T1-weighted imaging (T1WI) and hyperintense signal on T2-weighted imaging (T2WI) compared with extraocular muscles. Five cases had heterogeneous signals with focal areas of increased signal on T1WI or decreased signal on T2WI, including 1 case with hy-pointense signal on SWI..The mean apparent diffusion coeffi-cient (ADC) value of the viable part of tumors was (0.925 ± 0.09)×10-3 mm2/s. All cases showed moderate to marked en-hancement after contrast administration. Conclusion:Several MRI features-including homogeneous isointense or slightly hypointense signal on T1WI and slightly hyperintense signal on T2WI, relative low ADC values, and moderate to marked enhancement,.extraocular muscles inva-sion, and extraorbital extensionare helpful in the diagnosis of orbital RMS. (Eye Science 2014; 29:6-11).

  10. Infective endocarditis: the specific features of its course, the criteria for diagnosis, differential diagnosis (part II

    B S Belov


    Full Text Available Infective endocarditis (IE is today characterized by polyetiology due to a wide range of pathogens. The paper describes the specific features of the clinical picture of the disease in relation to the etiological agent, which have, in some cases, a crucial role in the choice of empiric antibiotic therapy. Significant clinical polymorphism, obscure symptoms, and monosyndromic onset as guises all enhance the importance of the differential diagnosis of IE, at its early stages in particular. Basic approaches to differentiating IE from the diseases in which differentially diagnostic problems arise to the utmost are outlined.

  11. [Clinical and imaging features of abdominal rhabdomyosarcoma of non-organ origin in children].

    Shi, J; Du, J; Wu, W; Wang, Q


    Objective: To evaluate the clinical and imaging features of abdominal rhabdomyosarcoma of non-organ origin in children. Methods: We retrospectively analyzed the clinical and imaging features of 12 pediatric patients with abdominal rhabdomyosarcoma confirmed by surgery and pathology at our hospital. Results: There were 9 boys and 3 girls, with an average age of (5.47±3.92) years old (range, 1 to 15). According to Intergroup Rhabdomyosarcoma Study (IRS) staging system, they were of stage Ⅲ to stage Ⅳ, and most were of embryonal type. Tumors of 7 cases were located in the pelvic cavity, 2 cases in the abdominal cavity, 1 in the retroperitoneal space, 1 in both the abdominal and pelvic cavities and 1 across the retroperitoneal space, and abdominal and pelvic cavities. Gray-scale ultrasound showed moderate inhomogeneous echo structure and color Doppler flow imaging showed rich blood flow signals. CT plain scan showed masses of iso- or low-density, and the contrast-enhanced scan showed lesions with inhomogeneous enhancement. The enhancement in delay scan was more obvious and the peripheral enhancement was more significant than central enhancement. Conclusions: Childhood abdominal rhabdomyosarcoma of non-organ origin may arise from the peritoneum, be commonly seen in boys younger than 10 years old, more likely located in the pelvic cavity, and embryonal rhabdomyosarcoma is the most common histological variant seen in childhood. Ultrasound and CT imaging can provide useful information for diagnosis and differential diagnosis of this tumor.

  12. A chondroblastoma versus a giant cell tumor: emphasis on the MR imaging features

    Chai, Jee Won; Hong, Sung Hwan; Choi, Ja Young; Kim, Na Ra; Choi, Jung Ah; Kang, Heung Sik [Seoul National University College of Medicine, Seoul (Korea, Republic of)


    To assess the MR imaging features in differentiating a chondroblastoma (CB) from a giant cell tumor (GCT), with an emphasis on the accompanying peritumoral bone marrow edema. MR imaging findings in 20 patients with CB were compared with the imaging features of 22 patients with GCT. The location of the lesion, signal intensity, adjacent cortical change, degree of accompanying bone marrow edema, synovitis in the adjacent joint and cystic change were analyzed. The findings of CB and GCT were examined statistically with use of Fisher's exact test. The incidence ratios of MR imaging findings were as follows (CB:GCT). Metaphyseal dominant involvement (2:21), partial cortical disruption (2:14), extensive bone marrow edema surrounding the tumor (14:0) and synovitis in the adjacent joint (11:2) were statistically different in incidence between CB and GCT ({rho} < 0.01). The inhomogeneous signal intensity (17:17) and cystic change (10:15) were not different in incidence between a CB and GCT. The presence of metaphyseal dominant involvement and cortical disruption favors a diagnosis of a GCT rather than a CB. In contrast, extensive bone marrow edema surrounding the tumor and synovitis in the adjacent joint are highly indicative of a CB.

  13. An image-tracking algorithm based on object center distance-weighting and image feature recognition

    JIANG Shuhong; WANG Qin; ZHANG Jianqiu; HU Bo


    Areal-time image-tracking algorithm is proposed.which gives small weights to pixels farther from the object center and uses the quantized image gray scales as a template.It identifies the target's location by the mean-shift iteration method and arrives at the target's scale by using image feature recognition.It improves the kernel-based algorithm in tracking scale-changing targets.A decimation mcthod is proposed to track large-sized targets and real-time experimental results verify the effectiveness of the proposed algorithm.

  14. Imaging trace element distributions in single organelles and subcellular features

    Kashiv, Yoav; Austin, Jotham R.; Lai, Barry; Rose, Volker; Vogt, Stefan; El-Muayed, Malek


    The distributions of chemical elements within cells are of prime importance in a wide range of basic and applied biochemical research. An example is the role of the subcellular Zn distribution in Zn homeostasis in insulin producing pancreatic beta cells and the development of type 2 diabetes mellitus. We combined transmission electron microscopy with micro- and nano-synchrotron X-ray fluorescence to image unequivocally for the first time, to the best of our knowledge, the natural elemental distributions, including those of trace elements, in single organelles and other subcellular features. Detected elements include Cl, K, Ca, Co, Ni, Cu, Zn and Cd (which some cells were supplemented with). Cell samples were prepared by a technique that minimally affects the natural elemental concentrations and distributions, and without using fluorescent indicators. It could likely be applied to all cell types and provide new biochemical insights at the single organelle level not available from organelle population level studies.

  15. Reversible acute methotrexate leukoencephalopathy: atypical brain MR imaging features

    Ziereisen, France; Damry, Nash; Christophe, Catherine [Queen Fabiola Children' s University Hospital, Department of Radiology, Brussels (Belgium); Dan, Bernard [Queen Fabiola Children' s University Hospital, Department of Neurology, Brussels (Belgium); Azzi, Nadira; Ferster, Alina [Queen Fabiola Children' s University Hospital, Department of Paediatrics, Brussels (Belgium)


    Unusual acute symptomatic and reversible early-delayed leukoencephalopathy has been reported to be induced by methotrexate (MTX). We aimed to identify the occurrence of such atypical MTX neurotoxicity in children and document its MR presentation. We retrospectively reviewed the clinical findings and brain MRI obtained in 90 children treated with MTX for acute lymphoblastic leukaemia or non-B malignant non-Hodgkin lymphoma. All 90 patients had normal brain imaging before treatment. In these patients, brain imaging was performed after treatment completion and/or relapse and/or occurrence of neurological symptoms. Of the 90 patients, 15 (16.7%) showed signs of MTX neurotoxicity on brain MRI, 9 (10%) were asymptomatic, and 6 (6.7%) showed signs of acute leukoencephalopathy. On the routine brain MRI performed at the end of treatment, all asymptomatic patients had classical MR findings of reversible MTX neurotoxicity, such as abnormal high-intensity areas localized in the deep periventricular white matter on T2-weighted images. In contrast, the six symptomatic patients had atypical brain MRI characterized by T2 high-intensity areas in the supratentorial cortex and subcortical white matter (n=6), cerebellar cortex and white matter (n=4), deep periventricular white matter (n=2) and thalamus (n=1). MR normalization occurred later than clinical recovery in these six patients. In addition to mostly asymptomatic classical MTX neurotoxicity, MTX may induce severe but reversible unusual leukoencephalopathy. It is important to recognize this clinicoradiological presentation in the differential diagnosis of acute neurological deterioration in children treated with MTX. (orig.)

  16. Histological Image Feature Mining Reveals Emergent Diagnostic Properties for Renal Cancer.

    Kothari, Sonal; Phan, John H; Young, Andrew N; Wang, May D


    Computer-aided histological image classification systems are important for making objective and timely cancer diagnostic decisions. These systems use combinations of image features that quantify a variety of image properties. Because researchers tend to validate their diagnostic systems on specific cancer endpoints, it is difficult to predict which image features will perform well given a new cancer endpoint. In this paper, we define a comprehensive set of common image features (consisting of 12 distinct feature subsets) that quantify a variety of image properties. We use a data-mining approach to determine which feature subsets and image properties emerge as part of an "optimal" diagnostic model when applied to specific cancer endpoints. Our goal is to assess the performance of such comprehensive image feature sets for application to a wide variety of diagnostic problems. We perform this study on 12 endpoints including 6 renal tumor subtype endpoints and 6 renal cancer grade endpoints. Keywords-histology, image mining, computer-aided diagnosis.

  17. Hemicrania Continua: Functional Imaging and Clinical Features With Diagnostic Implications.

    Sahler, Kristen


    This review focuses on summarizing 2 pivotal articles in the clinical and pathophysiologic understanding of hemicrania continua (HC). The first article, a functional imaging project, identifies both the dorsal rostral pons (a region associated with the generation of migraines) and the posterior hypothalamus (a region associated with the generation of cluster and short-lasting unilateral neuralgiform headache with conjunctival injection and tearing [SUNCT]) as active during HC. The second article is a summary of the clinical features seen in a prospective cohort of HC patients that carry significant diagnostic implications. In particular, they identify a wider range of autonomic signs than what is currently included in the International Headache Society criteria (including an absence of autonomic signs in a small percentage of patients), a high frequency of migrainous features, and the presence of aggravation and/or restlessness during attacks. Wide variations in exacerbation length, frequency, pain description, and pain location (including side-switching pain) are also noted. Thus, a case is made for widening and modifying the clinical diagnostic criteria used to identify patients with HC.

  18. Automated Image Retrieval of Chest CT Images Based on Local Grey Scale Invariant Features.

    Arrais Porto, Marcelo; Cordeiro d'Ornellas, Marcos


    Textual-based tools are regularly employed to retrieve medical images for reading and interpretation using current retrieval Picture Archiving and Communication Systems (PACS) but pose some drawbacks. All-purpose content-based image retrieval (CBIR) systems are limited when dealing with medical images and do not fit well into PACS workflow and clinical practice. This paper presents an automated image retrieval approach for chest CT images based local grey scale invariant features from a local database. Performance was measured in terms of precision and recall, average retrieval precision (ARP), and average retrieval rate (ARR). Preliminary results have shown the effectiveness of the proposed approach. The prototype is also a useful tool for radiology research and education, providing valuable information to the medical and broader healthcare community.

  19. Primary diaphyseal osteosarcoma in long bones: Imaging features and tumor characteristics

    Wang, Cheng-Sheng, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China); Yin, Qi-Hua, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China); Liao, Jin-Sheng, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China); Lou, Jiang-Hua, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China); Ding, Xiao-Yi, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China); Zhu, Yan-Bo, E-mail: [Department of Pathology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No.197, Ruijin 2nd Road, Shanghai 200025 (China); Chen, Ke-Min, E-mail: [Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No. 197, Ruijin 2nd Road, Shanghai 200025 (China)


    Objective: This study aims to assess retrospectively the imaging features of diaphyseal osteosarcoma and compare its characteristics with that of metaphyseal osteosarcoma. Materials and methods: Eighteen pathologically confirmed diaphyseal osteosarcomas were reviewed. Images of X-ray (n = 18), CT (n = 12) and MRI (n = 15) were evaluated by two radiologists. Differences among common radiologic findings of X-ray, CT and MRI, and between diaphyseal osteosarcomas and metaphyseal osteosarcomas in terms of tumor characteristics were compared. Results: The common imaging features of diaphyseal osteosarcoma were bone destruction, lamellar periosteal reaction with/without Codman triangle, massive soft tissue mass/swelling, neoplastic bone and/or calcification. CT and MRI had a higher detection rate in detecting bone destruction (P = 0.001) as compared with that of X-ray. X-ray and CT resulted in a higher percentage in detecting periosteal reaction (P = 0.018) and neoplastic bone and/or calcification (P = 0.043) as compared with that of MRI. There was no difference (P = 0.179) in detecting soft tissue mass among three imaging modalities. When comparing metaphyseal osteosarcoma to diaphyseal osteosarcoma, the latter had the following characteristics: a higher age of onset (P = 0.022), a larger extent of tumor (P = 0.018), a more osteolytic radiographic pattern (P = 0.043). Conclusion: As compared with metaphyseal osteosarcoma, diaphysial osteosarcoma is a special location of osteosarcoma with a lower incidence, a higher age of onset, a larger extent of tumor, a more osteolytic radiographic pattern. The osteoblastic and mixed types are diagnosed easily, but the osteolytic lesion should be differentiated from Ewing sarcoma. X-ray, CT and MRI can show imaging features from different aspects with different detection rates.

  20. Automatic solar feature detection using image processing and pattern recognition techniques

    Qu, Ming

    The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on Halpha full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves

  1. Plexiform Neurofibroma of the Wrist: Imaging Features and When to Suspect Malignancy

    Maria Gosein


    Full Text Available Plexiform neurofibromas are essentially pathognomonic for neurofibromatosis type 1 (NF1, occurring when there is diffuse involvement along a nerve segment and its branches. Transformation into a malignant peripheral nerve sheath tumour (MPNST is a major cause of mortality in NF1 patients. These tumours are highly aggressive and particularly difficult to diagnose in NF1 patients due to the clinical overlap between benign and malignant lesions. We present a case of a plexiform neurofibroma and discuss the typical imaging characteristics on ultrasound, CT, and MRI, including the target sign and continuity with the parent nerve. Certain imaging features should raise suspicion for malignancy however, these modalities may not always reliably differentiate between benign and malignant lesions. Recent studies show a very high negative predictive value for FDG-PET making it quite useful in excluding malignancy. In positive scans, PET/CT aids in guiding biopsy to the most metabolically active area of the tumour.

  2. Statistical analysis of textural features for improved classification of oral histopathological images.

    Muthu Rama Krishnan, M; Shah, Pratik; Chakraborty, Chandan; Ray, Ajoy K


    The objective of this paper is to provide an improved technique, which can assist oncopathologists in correct screening of oral precancerous conditions specially oral submucous fibrosis (OSF) with significant accuracy on the basis of collagen fibres in the sub-epithelial connective tissue. The proposed scheme is composed of collagen fibres segmentation, its textural feature extraction and selection, screening perfomance enhancement under Gaussian transformation and finally classification. In this study, collagen fibres are segmented on R,G,B color channels using back-probagation neural network from 60 normal and 59 OSF histological images followed by histogram specification for reducing the stain intensity variation. Henceforth, textural features of collgen area are extracted using fractal approaches viz., differential box counting and brownian motion curve . Feature selection is done using Kullback-Leibler (KL) divergence criterion and the screening performance is evaluated based on various statistical tests to conform Gaussian nature. Here, the screening performance is enhanced under Gaussian transformation of the non-Gaussian features using hybrid distribution. Moreover, the routine screening is designed based on two statistical classifiers viz., Bayesian classification and support vector machines (SVM) to classify normal and OSF. It is observed that SVM with linear kernel function provides better classification accuracy (91.64%) as compared to Bayesian classifier. The addition of fractal features of collagen under Gaussian transformation improves Bayesian classifier's performance from 80.69% to 90.75%. Results are here studied and discussed.

  3. Feature Extraction in Sequential Multimedia Images: with Applications in Satellite Images and On-line Videos

    Liang, Yu-Li

    Multimedia data is increasingly important in scientific discovery and people's daily lives. Content of massive multimedia is often diverse and noisy, and motion between frames is sometimes crucial in analyzing those data. Among all, still images and videos are commonly used formats. Images are compact in size but do not contain motion information. Videos record motion but are sometimes too big to be analyzed. Sequential images, which are a set of continuous images with low frame rate, stand out because they are smaller than videos and still maintain motion information. This thesis investigates features in different types of noisy sequential images, and the proposed solutions that intelligently combined multiple features to successfully retrieve visual information from on-line videos and cloudy satellite images. The first task is detecting supraglacial lakes above ice sheet in sequential satellite images. The dynamics of supraglacial lakes on the Greenland ice sheet deeply affect glacier movement, which is directly related to sea level rise and global environment change. Detecting lakes above ice is suffering from diverse image qualities and unexpected clouds. A new method is proposed to efficiently extract prominent lake candidates with irregular shapes, heterogeneous backgrounds, and in cloudy images. The proposed system fully automatize the procedure that track lakes with high accuracy. We further cooperated with geoscientists to examine the tracked lakes and found new scientific findings. The second one is detecting obscene content in on-line video chat services, such as Chatroulette, that randomly match pairs of users in video chat sessions. A big problem encountered in such systems is the presence of flashers and obscene content. Because of various obscene content and unstable qualities of videos capture by home web-camera, detecting misbehaving users is a highly challenging task. We propose SafeVchat, which is the first solution that achieves satisfactory

  4. A New Method of Semantic Feature Extraction for Medical Images Data

    XIE Conghua; SONG Yuqing; CHANG Jinyi


    In order to overcome the disadvantages of color, shape and texture-based features definition for medical images, this paper defines a new kind of semantic feature and its extraction algorithm. We firstly use kernel density estimation statistical model to describe the complicated medical image data, secondly, define some typical representative pixels of images as feature and finally, take hill-climbing strategy of Artificial Intelligence to extract those semantic features. Results of a content-based medial image retrieve system show that our semantic features have better distinguishing ability than those color, shape and texture-based features and can improve the ratios of recall and precision of this system smartly.

  5. Evaluation of feature selection algorithms for classification in temporal lobe epilepsy based on MR images

    Lai, Chunren; Guo, Shengwen; Cheng, Lina; Wang, Wensheng; Wu, Kai


    It's very important to differentiate the temporal lobe epilepsy (TLE) patients from healthy people and localize the abnormal brain regions of the TLE patients. The cortical features and changes can reveal the unique anatomical patterns of brain regions from the structural MR images. In this study, structural MR images from 28 normal controls (NC), 18 left TLE (LTLE), and 21 right TLE (RTLE) were acquired, and four types of cortical feature, namely cortical thickness (CTh), cortical surface area (CSA), gray matter volume (GMV), and mean curvature (MCu), were explored for discriminative analysis. Three feature selection methods, the independent sample t-test filtering, the sparse-constrained dimensionality reduction model (SCDRM), and the support vector machine-recursive feature elimination (SVM-RFE), were investigated to extract dominant regions with significant differences among the compared groups for classification using the SVM classifier. The results showed that the SVM-REF achieved the highest performance (most classifications with more than 92% accuracy), followed by the SCDRM, and the t-test. Especially, the surface area and gray volume matter exhibited prominent discriminative ability, and the performance of the SVM was improved significantly when the four cortical features were combined. Additionally, the dominant regions with higher classification weights were mainly located in temporal and frontal lobe, including the inferior temporal, entorhinal cortex, fusiform, parahippocampal cortex, middle frontal and frontal pole. It was demonstrated that the cortical features provided effective information to determine the abnormal anatomical pattern and the proposed method has the potential to improve the clinical diagnosis of the TLE.

  6. Computer-aided classification of Alzheimer's disease based on support vector machine with combination of cerebral image features in MRI

    Jongkreangkrai, C.; Vichianin, Y.; Tocharoenchai, C.; Arimura, H.; Alzheimer's Disease Neuroimaging Initiative


    Several studies have differentiated Alzheimer's disease (AD) using cerebral image features derived from MR brain images. In this study, we were interested in combining hippocampus and amygdala volumes and entorhinal cortex thickness to improve the performance of AD differentiation. Thus, our objective was to investigate the useful features obtained from MRI for classification of AD patients using support vector machine (SVM). T1-weighted MR brain images of 100 AD patients and 100 normal subjects were processed using FreeSurfer software to measure hippocampus and amygdala volumes and entorhinal cortex thicknesses in both brain hemispheres. Relative volumes of hippocampus and amygdala were calculated to correct variation in individual head size. SVM was employed with five combinations of features (H: hippocampus relative volumes, A: amygdala relative volumes, E: entorhinal cortex thicknesses, HA: hippocampus and amygdala relative volumes and ALL: all features). Receiver operating characteristic (ROC) analysis was used to evaluate the method. AUC values of five combinations were 0.8575 (H), 0.8374 (A), 0.8422 (E), 0.8631 (HA) and 0.8906 (ALL). Although “ALL” provided the highest AUC, there were no statistically significant differences among them except for “A” feature. Our results showed that all suggested features may be feasible for computer-aided classification of AD patients.

  7. Feature Based Image Mosaic Using Steerable Filters and Harris Corner Detector



    Full Text Available Image mosaic is to be combine several views of a scene in to single wide angle view. This paper proposes the feature based image mosaic approach. The mosaic image system includes feature point detection, feature point descriptor extraction and matching. A RANSAC algorithm is applied to eliminate number of mismatches and obtain transformation matrix between the images. The input image is transformed with the correct mapping model for image stitching and same is estimated. In this paper, feature points are detected using steerable filters and Harris, and compared with traditional Harris, KLT, and FAST corner detectors.

  8. Multi-scale contrast enhancement of oriented features in 2D images using directional morphology

    Das, Debashis; Mukhopadhyay, Susanta; Praveen, S. R. Sai


    This paper presents a multi-scale contrast enhancement scheme for improving the visual quality of directional features present in 2D gray scale images. Directional morphological filters are employed to locate and extract the scale-specific image features with different orientations which are subsequently stored in a set of feature images. The final enhanced image is constructed by weighted combination of these feature images with the original image. While construction, the feature images corresponding to progressively smaller scales are made to have higher proportion of contribution through the use of progressively larger weights. The proposed method has been formulated, implemented and executed on a set of real 2D gray scale images with oriented features. The experimental results visually establish the efficacy of the method. The proposed method has been compared with other similar methods both on subjective and objective basis and the overall performance is found to be satisfactory.

  9. Spindle epithelial tumor with thymus-like differentiation of the thyroid gland: A case report with ultrasonography and CT features, cytological findings and histopathological results

    Kang, Dong Joo; Lee, Yoo Jin; Kim, Dong Wook; Jung, Soo Jin [Busan Paik Hospital, Inje University College of Medicine, Busan (Korea, Republic of)


    Spindle epithelial tumor with thymus-like differentiation (SETTLE) of the thyroid gland is a very rare tumor. It is believed to originate from ectopic thymus tissue within the thyroid gland or from branchial pouch remnants that differentiate along the thymic line. A few reports of SETTLE have been presented, but to the best of our knowledge, there is no case report in which detailed preoperative imaging features of SETTLE have been described. In addition, there are no case reports of SETTLE in Korean patients. Thus, we report a case of SETTLE with detailed preoperative ultrasonography and computed tomography features, cytological findings and histopathological results.

  10. Perfusion MR imaging for differentiation of benign and malignant meningiomas

    Zhang, Hao; Rodiger, Lars A.; Shen, Tianzhen; Miao, Jingtao; Oudkerk, Matthijs


    Introduction Our purpose was to determine whether perfusion MR imaging can be used to differentiate benign and malignant meningiomas on the basis of the differences in perfusion of tumor parenchyma and/or peritumoral edema. Methods A total of 33 patients with preoperative meningiomas (25 benign and

  11. Electrodynamics, Differential Forms and the Method of Images

    Low, Robert J.


    This paper gives a brief description of how Maxwell's equations are expressed in the language of differential forms and use this to provide an elegant demonstration of how the method of images (well known in electrostatics) also works for electrodynamics in the presence of an infinite plane conducting boundary. The paper should be accessible to an…

  12. Differential phase contrast X-ray imaging system and components

    Stutman, Daniel; Finkenthal, Michael


    A differential phase contrast X-ray imaging system includes an X-ray illumination system, a beam splitter arranged in an optical path of the X-ray illumination system, and a detection system arranged in an optical path to detect X-rays after passing through the beam splitter.

  13. Differential phase contrast X-ray imaging system and components

    Stutman, Daniel; Finkenthal, Michael


    A differential phase contrast X-ray imaging system includes an X-ray illumination system, a beam splitter arranged in an optical path of the X-ray illumination system, and a detection system arranged in an optical path to detect X-rays after passing through the beam splitter.

  14. Cavernous Sinus: A Comprehensive Review of its Anatomy, Pathologic Conditions, and Imaging Features.

    Bakan, A A; Alkan, A; Kurtcan, S; Aralaşmak, A; Tokdemir, S; Mehdi, E; Özdemir, H


    The purpose of this article was to review the anatomy of the cavernous sinus (CS), illustrate numerous lesions that can affect the CS, and emphasize the imaging characteristics for each lesion to further refine the differential diagnoses. The CS, notwithstanding its small size, contains a complicated and crucial network that consists of the carotid artery, the venous plexus, and cranial nerves. The wide-ranging types of pathologies that can involve the CS can be roughly classified as tumoral, congenital, infectious/inflammatory/granulomatous, and vascular. Conditions that affect the CS usually lead to symptoms that are similar to each other; thus, for diagnosis, imaging procedures are required. Radiological evaluations are also required to detect pre- and postoperative CS invasion. Magnetic resonance imaging, which can be supplemented with thin-section contrast-enhanced sequences, is the preferred imaging technique for evaluating the CS. For correct diagnosis of CS lesions and accurate evaluations of CS invasions, it is essential to carefully analyze the anatomical structures within the CS and to acquire precise knowledge about the imaging features of CS lesions, which may frequently overlap.

  15. Medical Imaging in Differentiating the Diabetic Charcot Foot from Osteomyelitis.

    Short, Daniel J; Zgonis, Thomas


    Diabetic Charcot neuroarthropathy (DCN) poses a great challenge to diagnose in the early stages and when plain radiographs do not depict any initial signs of osseous fragmentation or dislocation in a setting of a high clinical index of suspicion. Medical imaging, including magnetic resonance imaging, computed tomography, and advanced bone scintigraphy, has its own unique clinical indications when treating the DCN with or without concomitant osteomyelitis. This article reviews different clinical case scenarios for choosing the most accurate medical imaging in differentiating DCN from osteomyelitis. Copyright © 2016 Elsevier Inc. All rights reserved.

  16. Differentiation of true anophthalmia from clinical anophthalmia using neuroradiological imaging

    Ali; Riza; Cenk; Celebi; Hadi; Sasani


    Anophthalmia is a condition of the absence of an eye and the presence of a small eye within the orbit.It is associated with many known syndromes.Clinical findings,as well as imaging modalities and genetic analysis,are important in making the diagnosis.Imaging modalities are crucial scanning methods.Cryptophthalmos,cyclopia,synophthalmia and congenital cystic eye should be considered in differential diagnoses.We report two clinical anophthalmic siblings,emphasizing the importance of neuroradiological and orbital imaging findings in distinguishing true congenital anophthalmia from clinical anophthalmia.

  17. Mixed solid and cystic acoustic neuroma: MR features and differential diagnosis

    Denys, A. [Service de Neuroradiologie-CIERM Hopital de Bicetre, Univ. de Paris Sud, 78, 94 Kremlin-Bicetre (France); Duvoisin, B. [Service de Neuroradiologie-CIERM Hopital de Bicetre, Univ. de Paris Sud, 78, 94 Kremlin-Bicetre (France)]|[Dept. of Radiodiagnosis, University Hospital, Lausanne (Switzerland); Fernandes, J.G. [Service de Neuroradiologie-CIERM Hopital de Bicetre, Univ. de Paris Sud, 78, 94 Kremlin-Bicetre (France); Doyon, D. [Service de Neuroradiologie-CIERM Hopital de Bicetre, Univ. de Paris Sud, 78, 94 Kremlin-Bicetre (France)


    We present a very rare case of combined cystic and solid acoustic neuroma investigated by magnetic resonance imaging (MRI). This case illustrates the value of MRI in the characterization of tumours in the posterior cranial fossa, particularly acoustic neuromas, and its diagnostic impact in unusual situations. The differential diagnosis of cystic and mixed lesions in the cerebellopontine angle is discussed. (orig.)

  18. Automatic Classification of Normal and Cancer Lung CT Images Using Multiscale AM-FM Features.

    Magdy, Eman; Zayed, Nourhan; Fakhr, Mahmoud


    Computer-aided diagnostic (CAD) systems provide fast and reliable diagnosis for medical images. In this paper, CAD system is proposed to analyze and automatically segment the lungs and classify each lung into normal or cancer. Using 70 different patients' lung CT dataset, Wiener filtering on the original CT images is applied firstly as a preprocessing step. Secondly, we combine histogram analysis with thresholding and morphological operations to segment the lung regions and extract each lung separately. Amplitude-Modulation Frequency-Modulation (AM-FM) method thirdly, has been used to extract features for ROIs. Then, the significant AM-FM features have been selected using Partial Least Squares Regression (PLSR) for classification step. Finally, K-nearest neighbour (KNN), support vector machine (SVM), naïve Bayes, and linear classifiers have been used with the selected AM-FM features. The performance of each classifier in terms of accuracy, sensitivity, and specificity is evaluated. The results indicate that our proposed CAD system succeeded to differentiate between normal and cancer lungs and achieved 95% accuracy in case of the linear classifier.

  19. Two-level evaluation on sensor interoperability of features in fingerprint image segmentation.

    Yang, Gongping; Li, Ying; Yin, Yilong; Li, Ya-Shuo


    Features used in fingerprint segmentation significantly affect the segmentation performance. Various features exhibit different discriminating abilities on fingerprint images derived from different sensors. One feature which has better discriminating ability on images derived from a certain sensor may not adapt to segment images derived from other sensors. This degrades the segmentation performance. This paper empirically analyzes the sensor interoperability problem of segmentation feature, which refers to the feature's ability to adapt to the raw fingerprints captured by different sensors. To address this issue, this paper presents a two-level feature evaluation method, including the first level feature evaluation based on segmentation error rate and the second level feature evaluation based on decision tree. The proposed method is performed on a number of fingerprint databases which are obtained from various sensors. Experimental results show that the proposed method can effectively evaluate the sensor interoperability of features, and the features with good evaluation results acquire better segmentation accuracies of images originating from different sensors.

  20. Optical Imaging for Stem Cell Differentiation to Neuronal Lineage

    Hwang, Do Won; Lee, Dong Soo [Seoul National Univ., Seoul (Korea, Republic of)


    In regenerative medicine, the prospect of stem cell therapy hold great promise for the recovery of injured tissues and effective treatment of intractable diseases. Tracking stem cell fate provides critical information to understand and evaluate the success of stem cell therapy. The recent emergence of in vivo noninvasive molecular imaging has enabled assessment of the behavior of grafted stem cells in living subjects. In this review, we provide an overview of current optical imaging strategies based on cell or tissue specific reporter gene expression and of in vivo methods to monitor stem cell differentiation into neuronal lineages. These methods use optical reporters either regulated by neuron-specific promoters or containing neuron-specific microRNA binding sites. Both systems revealed dramatic changes in optical reporter imaging signals in cells differentiating a yeast GAL4 amplification system or an engineering-enhanced luciferase reported gene. Furthermore, we propose an advanced imaging system to monitor neuronal differentiation during neurogenesis that uses in vivo multiplexed imaging techniques capable of detecting several targets simultaneously.

  1. Optical imaging for stem cell differentiation to neuronal lineage.

    Hwang, Do Won; Lee, Dong Soo


    In regenerative medicine, the prospect of stem cell therapy holds great promise for the recovery of injured tissues and effective treatment of intractable diseases. Tracking stem cell fate provides critical information to understand and evaluate the success of stem cell therapy. The recent emergence of in vivo noninvasive molecular imaging has enabled assessment of the behavior of grafted stem cells in living subjects. In this review, we provide an overview of current optical imaging strategies based on cell- or tissue-specific reporter gene expression and of in vivo methods to monitor stem cell differentiation into neuronal lineages. These methods use optical reporters either regulated by neuron-specific promoters or containing neuron-specific microRNA binding sites. Both systems revealed dramatic changes in optical reporter imaging signals in cells differentiating into a neuronal lineage. The detection limit of weak promoters or reporter genes can be greatly enhanced by adopting a yeast GAL4 amplification system or an engineering-enhanced luciferase reporter gene. Furthermore, we propose an advanced imaging system to monitor neuronal differentiation during neurogenesis that uses in vivo multiplexed imaging techniques capable of detecting several targets simultaneously.

  2. Divided-aperture differential confocal fast-imaging microscopy

    Wang, Yun; Qiu, Lirong; Zhao, Xiangye; Zhao, Weiqian


    A new method, laser divided-aperture differential confocal microscopy (DDCM), is proposed to achieve high-resolution 3D imaging of microstructures of large-scale sample surfaces. This method uses a divided-aperture confocal structure to significantly improve the axial resolution of confocal microscopy and keep a long working distance simultaneously; uses two radically offset point detectors to achieve differential detection to further improve the axial response sensitivity and realize fast imaging of a large-scale sample surface with a big axial scan-step interval. Theoretical analyses and experimental results show that the DDCM can reach an axial resolution of 5 nm with a 3.1 mm working distance with a 3 times imaging speed of a confocal system with the same resolution.

  3. Automated local bright feature image analysis of nuclear proteindistribution identifies changes in tissue phenotype

    Knowles, David; Sudar, Damir; Bator, Carol; Bissell, Mina


    The organization of nuclear proteins is linked to cell and tissue phenotypes. When cells arrest proliferation, undergo apoptosis, or differentiate, the distribution of nuclear proteins changes. Conversely, forced alteration of the distribution of nuclear proteins modifies cell phenotype. Immunostaining and fluorescence microscopy have been critical for such findings. However, there is an increasing need for quantitative analysis of nuclear protein distribution to decipher epigenetic relationships between nuclear structure and cell phenotype, and to unravel the mechanisms linking nuclear structure and function. We have developed imaging methods to quantify the distribution of fluorescently-stained nuclear protein NuMA in different mammary phenotypes obtained using three-dimensional cell culture. Automated image segmentation of DAPI-stained nuclei was generated to isolate thousands of nuclei from three-dimensional confocal images. Prominent features of fluorescently-stained NuMA were detected using a novel local bright feature analysis technique, and their normalized spatial density calculated as a function of the distance from the nuclear perimeter to its center. The results revealed marked changes in the distribution of the density of NuMA bright features as non-neoplastic cells underwent phenotypically normal acinar morphogenesis. In contrast, we did not detect any reorganization of NuMA during the formation of tumor nodules by malignant cells. Importantly, the analysis also discriminated proliferating non-neoplastic cells from proliferating malignant cells, suggesting that these imaging methods are capable of identifying alterations linked not only to the proliferation status but also to the malignant character of cells. We believe that this quantitative analysis will have additional applications for classifying normal and pathological tissues.

  4. Image Encryption Using Differential Evolution Approach in Frequency Domain

    Hassan, Maaly Awad S; 10.5121/sipij.2011.2105


    This paper presents a new effective method for image encryption which employs magnitude and phase manipulation using Differential Evolution (DE) approach. The novelty of this work lies in deploying the concept of keyed discrete Fourier transform (DFT) followed by DE operations for encryption purpose. To this end, a secret key is shared between both encryption and decryption sides. Firstly two dimensional (2-D) keyed discrete Fourier transform is carried out on the original image to be encrypted. Secondly crossover is performed between two components of the encrypted image, which are selected based on Linear Feedback Shift Register (LFSR) index generator. Similarly, keyed mutation is performed on the real parts of a certain components selected based on LFSR index generator. The LFSR index generator initializes it seed with the shared secret key to ensure the security of the resulting indices. The process shuffles the positions of image pixels. A new image encryption scheme based on the DE approach is developed...

  5. Scalable High-Performance Image Registration Framework by Unsupervised Deep Feature Representations Learning.

    Wu, Guorong; Kim, Minjeong; Wang, Qian; Munsell, Brent C; Shen, Dinggang


    Feature selection is a critical step in deformable image registration. In particular, selecting the most discriminative features that accurately and concisely describe complex morphological patterns in image patches improves correspondence detection, which in turn improves image registration accuracy. Furthermore, since more and more imaging modalities are being invented to better identify morphological changes in medical imaging data, the development of deformable image registration method that scales well to new image modalities or new image applications with little to no human intervention would have a significant impact on the medical image analysis community. To address these concerns, a learning-based image registration framework is proposed that uses deep learning to discover compact and highly discriminative features upon observed imaging data. Specifically, the proposed feature selection method uses a convolutional stacked autoencoder to identify intrinsic deep feature representations in image patches. Since deep learning is an unsupervised learning method, no ground truth label knowledge is required. This makes the proposed feature selection method more flexible to new imaging modalities since feature representations can be directly learned from the observed imaging data in a very short amount of time. Using the LONI and ADNI imaging datasets, image registration performance was compared to two existing state-of-the-art deformable image registration methods that use handcrafted features. To demonstrate the scalability of the proposed image registration framework, image registration experiments were conducted on 7.0-T brain MR images. In all experiments, the results showed that the new image registration framework consistently demonstrated more accurate registration results when compared to state of the art.

  6. Scalable High Performance Image Registration Framework by Unsupervised Deep Feature Representations Learning

    Wu, Guorong; Kim, Minjeong; Wang, Qian; Munsell, Brent C.


    Feature selection is a critical step in deformable image registration. In particular, selecting the most discriminative features that accurately and concisely describe complex morphological patterns in image patches improves correspondence detection, which in turn improves image registration accuracy. Furthermore, since more and more imaging modalities are being invented to better identify morphological changes in medical imaging data,, the development of deformable image registration method that scales well to new image modalities or new image applications with little to no human intervention would have a significant impact on the medical image analysis community. To address these concerns, a learning-based image registration framework is proposed that uses deep learning to discover compact and highly discriminative features upon observed imaging data. Specifically, the proposed feature selection method uses a convolutional stacked auto-encoder to identify intrinsic deep feature representations in image patches. Since deep learning is an unsupervised learning method, no ground truth label knowledge is required. This makes the proposed feature selection method more flexible to new imaging modalities since feature representations can be directly learned from the observed imaging data in a very short amount of time. Using the LONI and ADNI imaging datasets, image registration performance was compared to two existing state-of-the-art deformable image registration methods that use handcrafted features. To demonstrate the scalability of the proposed image registration framework image registration experiments were conducted on 7.0-tesla brain MR images. In all experiments, the results showed the new image registration framework consistently demonstrated more accurate registration results when compared to state-of-the-art. PMID:26552069

  7. Image Processing Based Customized Image Editor and Gesture Controlled Embedded Robot Coupled with Voice Control Features

    Somnath Kar


    Full Text Available In modern sciences and technologies, images gain much broader scopes due to the ever growing importance of scientific visualization (of often large-scale complex scientific/experimental data like microarray data in genetic research, or real-time multi-asset portfolio trading in finance etc. In this paper, a proposal has been presented to implement a Graphical User Interface (GUI consisting of various MATLAB functions related to image processing and using the same to create a basic image processing editor having different features like, viewing the red, green and blue components of a color image separately, color detection and various other features like noise addition and removal, edge detection, cropping, resizing, rotation, histogram adjust, brightness control that is used in a basic image editor along with object detection and tracking. This has been further extended to provide reliable and a more natural technique for the user to navigate a robot in the natural environment using gestures based on color tracking. Additionally, Voice control technique has been employed to navigate the robot in various directions in the Cartesian plane employing normal Speech recognition techniques available in Microsoft Visual Basic.


    M. Umemura


    Full Text Available We propose an image labeling method for LIDAR intensity image obtained by Mobile Mapping System (MMS using K-Nearest Neighbor (KNN of feature obtained by Convolutional Neural Network (CNN. Image labeling assigns labels (e.g., road, cross-walk and road shoulder to semantic regions in an image. Since CNN is effective for various image recognition tasks, we try to use the feature of CNN (Caffenet pre-trained by ImageNet. We use 4,096-dimensional feature at fc7 layer in the Caffenet as the descriptor of a region because the feature at fc7 layer has effective information for object classification. We extract the feature by the Caffenet from regions cropped from images. Since the similarity between features reflects the similarity of contents of regions, we can select top K similar regions cropped from training samples with a test region. Since regions in training images have manually-annotated ground truth labels, we vote the labels attached to top K similar regions to the test region. The class label with the maximum vote is assigned to each pixel in the test image. In experiments, we use 36 LIDAR intensity images with ground truth labels. We divide 36 images into training (28 images and test sets (8 images. We use class average accuracy and pixel-wise accuracy as evaluation measures. Our method was able to assign the same label as human beings in 97.8% of the pixels in test LIDAR intensity images.

  9. Image Labeling for LIDAR Intensity Image Using K-Nn of Feature Obtained by Convolutional Neural Network

    Umemura, Masaki; Hotta, Kazuhiro; Nonaka, Hideki; Oda, Kazuo


    We propose an image labeling method for LIDAR intensity image obtained by Mobile Mapping System (MMS) using K-Nearest Neighbor (KNN) of feature obtained by Convolutional Neural Network (CNN). Image labeling assigns labels (e.g., road, cross-walk and road shoulder) to semantic regions in an image. Since CNN is effective for various image recognition tasks, we try to use the feature of CNN (Caffenet) pre-trained by ImageNet. We use 4,096-dimensional feature at fc7 layer in the Caffenet as the descriptor of a region because the feature at fc7 layer has effective information for object classification. We extract the feature by the Caffenet from regions cropped from images. Since the similarity between features reflects the similarity of contents of regions, we can select top K similar regions cropped from training samples with a test region. Since regions in training images have manually-annotated ground truth labels, we vote the labels attached to top K similar regions to the test region. The class label with the maximum vote is assigned to each pixel in the test image. In experiments, we use 36 LIDAR intensity images with ground truth labels. We divide 36 images into training (28 images) and test sets (8 images). We use class average accuracy and pixel-wise accuracy as evaluation measures. Our method was able to assign the same label as human beings in 97.8% of the pixels in test LIDAR intensity images.

  10. Magnetic resonance imaging features of asymptomatic bipartite patella

    O' Brien, J., E-mail: [Department of Radiology, Adelaide and Meath Incorporating National Children' s Hospital, Tallaght, Dublin 24 (Ireland); Murphy, C.; Halpenny, D.; McNeill, G.; Torreggiani, W.C. [Department of Radiology, Adelaide and Meath Incorporating National Children' s Hospital, Tallaght, Dublin 24 (Ireland)


    Objective: The purpose of our study was to describe the magnetic resonance imaging (MRI) features of bipartite patella in asymptomatic patients. Materials and methods: The study was prospective in type and performed following institutional ethical committees approval. In total, 25 subjects were recruited into the study and informed consent obtained in each case. The local radiology database was utilised in conjunction with a clinical questionnaire to identify patients who had asymptomatic bipartite patella. Any patient with a history of trauma or symptomatic disease was excluded from the study. MRI imaging was performed in each case on a 1.5 T system using a dedicated knee coil and a standardised knee protocol. The images obtained were then analysed by two musculoskeletal radiologists in consensus. Results: Of the 25 subjects, there were 8 females and 17 males. The mean age was 34.6 years. All but one of the bipartite fragments were located on the superolateral aspect of the patella. In 23 cases, one fragment was identified. The average transverse diameter of the fragment was 12.8 mm. The average distance between the fragment and the adjacent patella in the axial plane was 1.46 mm. In addition, the cartilage overlying the patella and accessory fragment was intact in all cases. The average thickness of the patella cartilage at its border to the fragment was 2.4 mm with an average ratio of the cartilage thickness of the fragment as compared with the cartilage thickness of the patella of 0.72. There was no evidence of high signal or bone marrow oedema on fluid sensitive sequences within either the patella or the fragment in any of the patients. Fluid was identified in the cleft between the patella and the fragment in the majority of cases. Conclusions: Asymptomatic bipartite patella is characterised by intact but thinned cartilage along the border between the patella and the fragment, fluid between the cleft and a lack of any bone marrow oedema or high signal within

  11. An Open Source Agenda for Research Linking Text and Image Content Features.

    Goodrum, Abby A.; Rorvig, Mark E.; Jeong, Ki-Tai; Suresh, Chitturi


    Proposes methods to utilize image primitives to support term assignment for image classification. Proposes to release code for image analysis in a common tool set for other researchers to use. Of particular focus is the expansion of work by researchers in image indexing to include image content-based feature extraction capabilities in their work.…

  12. Detection of Brain Tumor and Extraction of Texture Features using Magnetic Resonance Images

    Prof. Dilip Kumar Gandhi


    Full Text Available Brain Cancer Detection system is designed. Aim of this paper is to locate the tumor and determine the texture features from a Brain Cancer affected MRI. A computer based diagnosis is performed in order to detect the tumors from given Magnetic Resonance Image. Basic image processing techniques are used to locate the tumor region. Basic techniques consist of image enhancement, image bianarization, and image morphological operations. Texture features are computed using the Gray Level Co-occurrence Matrix. Texture features consists of five distinct features. Selective features or the combination of selective features will be used in the future to determine the class of the query image. Astrocytoma type of Brain Cancer affected images are used only for simplicity

  13. Imaging features of central nervous system fungal infections

    Jain Krishan


    Full Text Available Fungal infections of the central nervous system (CNS are rare in the general population and are invariably secondary to primary focus elsewhere, usually in the lung or intestine. Except for people with longstanding diabetes, they are most frequently encountered in immunocompromised patients such as those with acquired immunodeficiency syndrome or after organ transplantation. Due to the lack of inflammatory response, neuroradiological findings are often nonspecific and are frequently mistaken for tuberculous meningitis, pyogenic abscess or brain tumor. Intracranial fungal infections are being identified more frequently due to the increased incidence of AIDS patients, better radiological investigations, more sensitive microbiological techniques and better critical care of moribund patients. Although almost any fungus may cause encephalitis, cryptococcal meningoencephalitis is most frequently seen, followed by aspergillosis and candidiasis. The biology, epidemiology and imaging features of the common fungal infections of the CNS will be reviewed. The radiographic appearance alone is often not specific, but the combination of the appropriate clinical setting along with computed tomography or magnetic resonance may help to suggest the correct diagnosis.


    Irianto .


    Full Text Available The central problem of most Content Based Image Retrieval approaches is poor quality in terms of sensitivity (recall and specificity (precision. To overcome this problem, the semantic gap between high-level concepts and low-level features has been acknowledged. In this paper we introduce an approach to reduce the impact of the semantic gap by integrating high-level (semantic and low-level features to improve the quality of Image Retrieval queries. Our experiments have been carried out by applying two hierarchical procedures. The first approach is called keyword-content, and the second content-keyword. Our proposed approaches show better results compared to a single method (keyword or content based in term of recall and precision. The average precision has increased by up to 50%.

  15. A Novel Feature Extraction Scheme for Medical X-Ray Images

    Prachi.G.Bhende; Dr.A.N.Cheeran


    X-ray images are gray scale images with almost the same textural characteristic. Conventional texture or color features cannot be used for appropriate categorization in medical x-ray image archives. This paper presents a novel combination of methods like GLCM, LBP and HOG for extracting distinctive invariant features from Xray images belonging to IRMA (Image Retrieval in Medical applications) database that can be used to perform reliable matching between different views of an obje...

  16. Magnetic resonance imaging differential diagnosis of brainstem lesions in children

    Carlo Cosimo Quattrocchi; Yuri Errante; Maria Camilla Rossi Espagnet; Stefania Galassi; Sabino Walter Della Sala; Bruno Bernardi; Giuseppe Fariello; Daniela Longo


    Differential diagnosis of brainstem lesions,either isolated or in association with cerebellar and supra-tentorial lesions,can be challenging. Knowledge of the structural organization is crucial for the differential diagnosis and establishment of prognosis of pathologies with involvement of the brainstem. Familiarity with the location of the lesions in the brainstem is essential,especially in the pediatric population. Magnetic resonance imaging(MRI) is the most sensitive and specific imaging technique for diagnosing disorders of the posterior fossa and,particularly,the brainstem. High magnetic static field MRI allows detailed visualization of the morphology,signal intensity and metabolic content of the brainstem nuclei,together with visualization of the normal development and myelination. In this pictorial essay we review the brainstem pathology in pediatric patients and consider the MR imaging patterns that may help the radiologist to differentiate among vascular,toxico-metabolic,infectiveinflammatory,degenerative and neoplastic processes. Helpful MR tips can guide the differential diagnosis: These include the location and morphology of lesions,the brainstem vascularization territories,gray and white matter distribution and tissue selective vulnerability.

  17. The shape operator for differential analysis of images.

    Avants, Brian; Gee, James


    This work provides a new technique for surface oriented volumetric image analysis. The method makes no assumptions about topology, instead constructing a local neighborhood from image information, such as a segmentation or edge map, to define a surface patch. Neighborhood constructions using extrinsic and intrinsic distances are given. This representation allows one to estimate differential properties directly from the image's Gauss map. We develop a novel technique for this purpose which estimates the shape operator and yields both principal directions and curvatures. Only first derivatives need be estimated, making the method numerically stable. We show the use of these measures for multi-scale classification of image structure by the mean and Gaussian curvatures. Finally, we propose to register image volumes by surface curvature. This is particularly useful when geometry is the only variable. To illustrate this, we register binary segmented data by surface curvature, both rigidly and non-rigidly. A novel variant of Demons registration, extensible for use with differentiable similarity metrics, is also applied for deformable curvature-driven registration of medical images.

  18. Feature-point-extracting-based automatically mosaic for composite microscopic images

    YIN YanSheng; ZHAO XiuYang; TIAN XiaoFeng; LI Jia


    Image mosaic is a crucial step in the three-dimensional reconstruction of composite materials to align the serial images. A novel method is adopted to mosaic two SiC/Al microscopic images with an amplification coefficient of 1000. The two images are denoised by Gaussian model, and feature points are then extracted by using Harris corner detector. The feature points are filtered through Canny edge detector. A 40x40 feature template is chosen by sowing a seed in an overlapped area of the reference image, and the homologous region in floating image is acquired automatically by the way of correlation analysis. The feature points in matched templates are used as feature point-sets. Using the transformational parameters acquired by SVD-ICP method, the two images are transformed into the universal coordinates and merged to the final mosaic image.

  19. Prediction of pancreatic neuroendocrine tumour grade with MR imaging features: added value of diffusion-weighted imaging

    Lotfalizadeh, Emad; Vullierme, Marie-Pierre; Allaham, Wassim [University Hospitals Paris Nord Val de Seine, Department of Radiology, Clichy, Hauts-de-Seine (France); Ronot, Maxime; Vilgrain, Valerie [University Hospitals Paris Nord Val de Seine, Department of Radiology, Clichy, Hauts-de-Seine (France); University Paris Diderot, Paris (France); INSERM U1149, Centre de Recherche Biomedicale Bichat-Beaujon, CRB3, Paris (France); Wagner, Mathilde [University Hospitals Paris Nord Val de Seine, Department of Radiology, Clichy, Hauts-de-Seine (France); INSERM U1149, Centre de Recherche Biomedicale Bichat-Beaujon, CRB3, Paris (France); Cros, Jerome; Couvelard, Anne [University Paris Diderot, Paris (France); University Hospitals Paris Nord Val de Seine, Department of Pathology, Clichy, Hauts-de-Seine (France); Hentic, Olivia; Ruzniewski, Philippe [University Hospitals Paris Nord Val de Seine, Department of Gastroenterology, Clichy, Hauts-de-Seine (France)


    To evaluate the value of MR imaging including diffusion-weighted imaging (DWI) for the grading of pancreatic neuroendocrine tumours (pNET). Between 2006 and 2014, all resected pNETs with preoperative MR imaging including DWI were included. Tumour grading was based on the 2010 WHO classification. MR imaging features included size, T1-w, and T2-w signal intensity, enhancement pattern, apparent (ADC) and true diffusion (D) coefficients. One hundred and eight pNETs (mean 40 ± 33 mm) were evaluated in 94 patients (48 women, 51 %, mean age 52 ± 12). Fifty-five (51 %), 42 (39 %), and 11 (10 %) tumours were given the following grades (G): G1, G2, and G3. Mean ADC and D values were significantly lower as grade increased (ADC: 2.13 ± 0.70, 1.78 ± 0.72, and 0.86 ± 0.22 10{sup -3} mm{sup 2}/s, and D: 1.92 ± 0.70, 1.75 ± 0.74, and 0.82 ± 0.19 10{sup -3} mm{sup 2}/s G1, G2, and G3, all p < 0.001). A higher grade was associated with larger sized tumours (p < 0.001). The AUROC of ADC and D to differentiate G3 and G1-2 were 0.96 ± 0.02 and 0.95 ± 0.02. Optimal cut-off values for the identification of G3 were 1.19 10{sup -3} mm{sup 2}/s for ADC (sensitivity 100 %, specificity 92 %) and 1.04 10{sup -3} mm{sup 2}/s for D (sensitivity 82 %, specificity 92 %). Morphological/functional MRI features of pNETS depend on tumour grade. DWI is useful for the identification of high-grade tumours. (orig.)

  20. Diabetic mastopathy: Imaging features and the role of image-guided biopsy in its diagnosis

    Kim, Jong Hyeon; Kim, Eun Kyung; Kim, Min Jung; Moon, Hee Jung; Yoon, Jung Hyun [Dept. of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul (Korea, Republic of)


    The goal of this study was to evaluate the imaging features of diabetic mastopathy (DMP) and the role of image-guided biopsy in its diagnosis. Two experienced radiologists retrospectively reviewed the mammographic and sonographic images of 19 pathologically confirmed DMP patients. The techniques and results of the biopsies performed in each patient were also reviewed. Mammograms showed negative findings in 78% of the patients. On ultrasonography (US), 13 lesions were seen as masses and six as non-mass lesions. The US features of the mass lesions were as follows: irregular shape (69%), oval shape (31%), indistinct margin (69%), angular margin (15%), microlobulated margin (8%), well-defined margin (8%), heterogeneous echogenicity (62%), hypoechoic echogenicity (38%), posterior shadowing (92%), parallel orientation (100%), the absence of calcifications (100%), and the absence of vascularity (100%). Based on the US findings, 17 lesions (89%) were classified as Breast Imaging Reporting and Data System category 4 and two (11%) as category 3. US-guided core biopsy was performed in 18 patients, and 10 (56%) were diagnosed with DMP on that basis. An additional vacuum-assisted biopsy was performed in seven patients and all were diagnosed with DMP. The US features of DMP were generally suspicious for malignancy, whereas the mammographic findings were often negative or showed only focal asymmetry. Core biopsy is an adequate method for initial pathological diagnosis. However, since it yields non-diagnostic results in a considerable number of cases, the evaluation of correlations between imaging and pathology plays an important role in the diagnostic process.

  1. Sciatica-like symptoms and the sacroiliac joint: clinical features and differential diagnosis

    Visser, L.H.; Nijssen, P.G.; Tijssen, C.C.; Middendorp, J.J. van; Schieving, J.H.


    PURPOSE: To compare the clinical features of patients with sacroiliac joint (SIJ)-related sciatica-like symptoms to those with sciatica from nerve root compression and to investigate the necessity to perform radiological imaging in patients with sciatica-like symptoms derived from the SIJ. METHODS:

  2. An Algorithm of Image Contrast Enhancement Based on Pixels Neighborhood’s Local Feature

    Chen Yan


    Full Text Available In this study, we proposed an algorithm of Image Contrast enhancement based on local feature to acquire edge information of image, remove Ray Imaging noise and overcome edge blurry and other defects. This method can extract edge features and finish contrast enhancement in varying degrees for pixels neighborhood with different characteristics by using neighborhood local variance and complexity function, which can achieve local features enhancement. The stimulation shows that the method can not only enhance the contrast of the entire image, but also effectively preserves image edge information and improve image quality.

  3. Topographic Feature Extraction for Bengali and Hindi Character Images

    Bag, Soumen; 10.5121/sipij.2011.2215


    Feature selection and extraction plays an important role in different classification based problems such as face recognition, signature verification, optical character recognition (OCR) etc. The performance of OCR highly depends on the proper selection and extraction of feature set. In this paper, we present novel features based on the topography of a character as visible from different viewing directions on a 2D plane. By topography of a character we mean the structural features of the strokes and their spatial relations. In this work we develop topographic features of strokes visible with respect to views from different directions (e.g. North, South, East, and West). We consider three types of topographic features: closed region, convexity of strokes, and straight line strokes. These features are represented as a shape-based graph which acts as an invariant feature set for discriminating very similar type characters efficiently. We have tested the proposed method on printed and handwritten Bengali and Hindi...

  4. The Application of Partial Differential Equations in Medical Image Processing

    Mohammad Madadpour Inallou


    Full Text Available Mathematical models are the foundation of biomedical computing. Partial Differential Equations (PDEs in Medical Imaging is concerned with acquiring images of the body for research, diagnosis and treatment. Biomedical Image Processing and its influence has undergoing a revolution in the past decade. Image processing has become an important component in contemporary science and technology and has been an interdisciplinary research field attracting expertise from applied mathematics, biology, computer sciences, engineering, statistics, microscopy, radiologic sciences, physics, medicine and etc. Medical imaging equipment is taking on an increasingly critical role in healthcare as the industry strives to lower patient costs and achieve earlier disease prediction using noninvasive means. The subsections of medical imaging are categorized to two: Conventional (X-Ray and Ultrasound and Computed (CT, MRI, fMRI, SPECT, PET and etc. This paper is organized as fallow: First section describes some kind of image processing. Second section is about techniques and requirements, and in the next sections the proceeding of Analyzing, Smoothing, Segmentation, De-noising and Registration in Medical Image Processing Equipment by PDEs Framework will be regarded

  5. Image mining and Automatic Feature extraction from Remotely Sensed Image (RSI using Cubical Distance Methods



    Full Text Available Information processing and decision support system using image mining techniques is in advance drive with huge availability of remote sensing image (RSI. RSI describes inherent properties of objects by recording their natural reflectance in the electro-magnetic spectral (ems region. Information on such objects could be gathered by their color properties or their spectral values in various ems range in the form of pixels. Present paper explains a method of such information extraction using cubical distance method and subsequent results. Thismethod is one among the simpler in its approach and considers grouping of pixels on the basis of equal distance from a specified point in the image or selected pixel having definite attribute values (DN in different spectral layers of the RSI. The color distance and the occurrence pixel distance play a vital role in determining similarobjects as clusters aid in extracting features in the RSI domain.

  6. Central nervous system tumors with ependymal features: a broadened spectrum of primarily ependymal differentiation?

    Lehman, Norman L


    Ependymomas are well-characterized central nervous system (CNS) tumors that occur most often in children and young adults. Several other CNS tumor entities, including astroblastoma, chordoid glioma, papillary tumor of the pineal region, angiocentric glioma, and pilomyxoid astrocytoma, variably display histopathologic features of ependymal differentiation. The ependymal differentiation in some of these tumors is generally accepted, whereas in others, it is controversial. This article briefly reviews ependymal cell development and conventional ependymomas, the pathologic findings and clinical behavior of tumors with variable ependymal features, and the rationales for their inclusion with ependymomas or exclusion from a larger family of ependymal tumors. These issues are addressed in the context of early morphologic insights of Bailey and Cushing, Friede, and others; contemporary oncologic concepts; and recent relevant molecular and tumor stem cell studies.

  7. Optimal method for exoplanet detection by angular differential imaging.

    Mugnier, Laurent M; Cornia, Alberto; Sauvage, Jean-François; Rousset, Gérard; Fusco, Thierry; Védrenne, Nicolas


    We propose a novel method for the efficient direct detection of exoplanets from the ground using angular differential imaging. The method combines images appropriately, then uses the combined images jointly in a maximum-likelihood framework to estimate the position and intensity of potential planets orbiting the observed star. It takes into account the mixture of photon and detector noises and a positivity constraint on the planet's intensity. A reasonable detection criterion is also proposed based on the computation of the noise propagation from the images to the estimated intensity of the potential planet. The implementation of this method is tested on simulated data that take into account static aberrations before and after the coronagraph, residual turbulence after adaptive optics correction, and noise.

  8. Pleasant/Unpleasant Filtering for Affective Image Retrieval Based on Cross-Correlation of EEG Features

    Keranmu Xielifuguli


    Full Text Available People often make decisions based on sensitivity rather than rationality. In the field of biological information processing, methods are available for analyzing biological information directly based on electroencephalogram: EEG to determine the pleasant/unpleasant reactions of users. In this study, we propose a sensitivity filtering technique for discriminating preferences (pleasant/unpleasant for images using a sensitivity image filtering system based on EEG. Using a set of images retrieved by similarity retrieval, we perform the sensitivity-based pleasant/unpleasant classification of images based on the affective features extracted from images with the maximum entropy method: MEM. In the present study, the affective features comprised cross-correlation features obtained from EEGs produced when an individual observed an image. However, it is difficult to measure the EEG when a subject visualizes an unknown image. Thus, we propose a solution where a linear regression method based on canonical correlation is used to estimate the cross-correlation features from image features. Experiments were conducted to evaluate the validity of sensitivity filtering compared with image similarity retrieval methods based on image features. We found that sensitivity filtering using color correlograms was suitable for the classification of preferred images, while sensitivity filtering using local binary patterns was suitable for the classification of unpleasant images. Moreover, sensitivity filtering using local binary patterns for unpleasant images had a 90% success rate. Thus, we conclude that the proposed method is efficient for filtering unpleasant images.

  9. Fake/Bogus Conferences: Their Features and Some Subtle Ways to Differentiate Them from Real Ones.

    Asadi, Amin; Rahbar, Nader; Rezvani, Mohammad Javad; Asadi, Fahime


    The main objective of the present paper is to introduce some features of fake/bogus conferences and some viable approaches to differentiate them from the real ones. These fake/bogus conferences introduce themselves as international conferences, which are multidisciplinary and indexed in major scientific digital libraries. Furthermore, most of the fake/bogus conference holders offer publishing the accepted papers in ISI journals and use other techniques in their advertisement e-mails.


    Yogapriya Jaganathan


    Full Text Available Feature dimensionality reduction problem is a major issue in Content Based Medical Image Retrieval (CBMIR for the effective management of medical images with the support of visual features for the purpose of diagnosis and educational research field. However, high dimensional features would be an origin for substantial challenges in retrieval. The proposed CBMIR is used a unified approach based on extraction of visual features, optimized feature selection, classification of optimized features and similarity measurements. However, high dimensional features would be an origin for substantial challenges in retrieval. The Texture features are selected using Gray Level Co-occurrence Matrix (GLCM, Tamura Features (TF and Gabor Filter (GF in which pull out of features are formed a feature vector database. Fuzzy based PSO (FPSO is applied for Feature selection to overcome the difficulty of feature vectors being surrounded in local optima of original PSO. This procedure also integrates a smart policymaking structure of ACO procedure into the novel FPSO where the global optimum position to be exclusive for every feature particle. The Fuzzy based Particle Swarm Optimization and Ant Colony Optimization (FPSO-ACO technique is used to trim down the feature vector dimensionality and classification is accomplished using an extensive Fuzzy based Relevance Vector Machine (FRVM to form collections of relevant image features that would provide an accepted way to classify dimensionally concentrated feature vectors of images. The Euclidean Distance (ED is recognized as finest for similarity measurement between the medical query image and the medical image database. This proposed approach can acquire the query from the user and had retrieved the desired images from the database. The retrieval performance would be assessed based on precision and recall. This proposed CBMIR is used to provide comfort to the physician to obtain more assurance in their decisions for

  11. Performance Evaluation of Content Based Image Retrieval on Feature Optimization and Selection Using Swarm Intelligence

    Kirti Jain


    Full Text Available The diversity and applicability of swarm intelligence is increasing everyday in the fields of science and engineering. Swarm intelligence gives the features of the dynamic features optimization concept. We have used swarm intelligence for the process of feature optimization and feature selection for content-based image retrieval. The performance of content-based image retrieval faced the problem of precision and recall. The value of precision and recall depends on the retrieval capacity of the image. The basic raw image content has visual features such as color, texture, shape and size. The partial feature extraction technique is based on geometric invariant function. Three swarm intelligence algorithms were used for the optimization of features: ant colony optimization, particle swarm optimization (PSO, and glowworm optimization algorithm. Coral image dataset and MatLab software were used for evaluating performance.

  12. Biomedical Imaging Modality Classification Using Combined Visual Features and Textual Terms

    Xian-Hua Han


    extraction from medical images and fuses the different extracted visual features and textual feature for modality classification. To extract visual features from the images, we used histogram descriptor of edge, gray, or color intensity and block-based variation as global features and SIFT histogram as local feature. For textual feature of image representation, the binary histogram of some predefined vocabulary words from image captions is used. Then, we combine the different features using normalized kernel functions for SVM classification. Furthermore, for some easy misclassified modality pairs such as CT and MR or PET and NM modalities, a local classifier is used for distinguishing samples in the pair modality to improve performance. The proposed strategy is evaluated with the provided modality dataset by ImageCLEF 2010.

  13. Conversion Discriminative Analysis on Mild Cognitive Impairment Using Multiple Cortical Features from MR Images

    Shengwen Guo


    Full Text Available Neuroimaging measurements derived from magnetic resonance imaging provide important information required for detecting changes related to the progression of mild cognitive impairment (MCI. Cortical features and changes play a crucial role in revealing unique anatomical patterns of brain regions, and further differentiate MCI patients from normal states. Four cortical features, namely, gray matter volume, cortical thickness, surface area, and mean curvature, were explored for discriminative analysis among three groups including the stable MCI (sMCI, the converted MCI (cMCI, and the normal control (NC groups. In this study, 158 subjects (72 NC, 46 sMCI, and 40 cMCI were selected from the Alzheimer's Disease Neuroimaging Initiative. A sparse-constrained regression model based on the l2-1-norm was introduced to reduce the feature dimensionality and retrieve essential features for the discrimination of the three groups by using a support vector machine (SVM. An optimized strategy of feature addition based on the weight of each feature was adopted for the SVM classifier in order to achieve the best classification performance. The baseline cortical features combined with the longitudinal measurements for 2 years of follow-up data yielded prominent classification results. In particular, the cortical thickness produced a classification with 98.84% accuracy, 97.5% sensitivity, and 100% specificity for the sMCI–cMCI comparison; 92.37% accuracy, 84.78% sensitivity, and 97.22% specificity for the cMCI–NC comparison; and 93.75% accuracy, 92.5% sensitivity, and 94.44% specificity for the sMCI–NC comparison. The best performances obtained by the SVM classifier using the essential features were 5–40% more than those using all of the retained features. The feasibility of the cortical features for the recognition of anatomical patterns was certified; thus, the proposed method has the potential to improve the clinical diagnosis of sub-types of MCI and

  14. Neuron Segmentation in Electron Microscopy Images Using Partial Differential Equations.

    Jones, Cory; Sayedhosseini, Mojtaba; Ellisman, Mark; Tasdizen, Tolga


    In connectomics, neuroscientists seek to identify the synaptic connections between neurons. Segmentation of cell membranes using supervised learning algorithms on electron microscopy images of brain tissue is often done to assist in this effort. Here we present a partial differential equation with a novel growth term to improve the results of a supervised learning algorithm. We also introduce a new method for representing the resulting image that allows for a more dynamic thresholding to further improve the result. Using these two processes we are able to close small to medium sized gaps in the cell membrane detection and improve the Rand error by as much as 9% over the initial supervised segmentation.

  15. MR imaging of the forefoot: Morton neuroma and differential diagnoses.

    Zanetti, Marco; Weishaupt, Dominik


    Magnetic resonance (MR) imaging of Morton neuromas is highly accurate. Morton neuromas are more conspicuous when the patient is prone positioned and the foot is plantar flexed than in the supine position with the toes pointing upward. MR imaging of Morton neuromas has a large influence on the diagnostic thinking and treatment plan of orthopedic foot surgeons. The most common differential diagnoses include intermetatarsal bursitis, stress fractures, and stress reactions. Some diagnoses (nodules associated with rheumatoid arthritis, synovial cyst, soft tissue chondroma, and plantar fibromatosis) are rare and can be diagnosed with histologic correlation only.

  16. Skin cancer texture analysis of OCT images based on Haralick, fractal dimension, Markov random field features, and the complex directional field features

    Raupov, Dmitry S.; Myakinin, Oleg O.; Bratchenko, Ivan A.; Zakharov, Valery P.; Khramov, Alexander G.


    In this paper, we propose a report about our examining of the validity of OCT in identifying changes using a skin cancer texture analysis compiled from Haralick texture features, fractal dimension, Markov random field method and the complex directional features from different tissues. Described features have been used to detect specific spatial characteristics, which can differentiate healthy tissue from diverse skin cancers in cross-section OCT images (B- and/or C-scans). In this work, we used an interval type-II fuzzy anisotropic diffusion algorithm for speckle noise reduction in OCT images. The Haralick texture features as contrast, correlation, energy, and homogeneity have been calculated in various directions. A box-counting method is performed to evaluate fractal dimension of skin probes. Markov random field have been used for the quality enhancing of the classifying. Additionally, we used the complex directional field calculated by the local gradient methodology to increase of the assessment quality of the diagnosis method. Our results demonstrate that these texture features may present helpful information to discriminate tumor from healthy tissue. The experimental data set contains 488 OCT-images with normal skin and tumors as Basal Cell Carcinoma (BCC), Malignant Melanoma (MM) and Nevus. All images were acquired from our laboratory SD-OCT setup based on broadband light source, delivering an output power of 20 mW at the central wavelength of 840 nm with a bandwidth of 25 nm. We obtained sensitivity about 97% and specificity about 73% for a task of discrimination between MM and Nevus.

  17. Time-Resolved Imaging of Negative Differential Resistance on the Atomic Scale

    Rashidi, Mohammad; Taucer, Marco; Ozfidan, Isil; Lloyd, Erika; Koleini, Mohammad; Labidi, Hatem; Pitters, Jason L.; Maciejko, Joseph; Wolkow, Robert A.


    Negative differential resistance remains an attractive but elusive functionality, so far only finding niche applications. Atom scale entities have shown promising properties, but the viability of device fabrication requires a fuller understanding of electron dynamics than has been possible to date. Using an all-electronic time-resolved scanning tunneling microscopy technique and a Green's function transport model, we study an isolated dangling bond on a hydrogen terminated silicon surface. A robust negative differential resistance feature is identified as a many body phenomenon related to occupation dependent electron capture by a single atomic level. We measure all the time constants involved in this process and present atomically resolved, nanosecond time scale images to simultaneously capture the spatial and temporal variation of the observed feature.

  18. Content-Based Digital Image Retrieval based on Multi-Feature Amalgamation

    Linhao Li


    Full Text Available In actual implementation, digital image retrieval are facing all kinds of problems. There still exists some difficulty in measures and methods for application. Currently there is not a unambiguous algorithm which can directly shown the obvious feature of image content and satisfy the color, scale invariance and rotation invariance of feature simultaneously. So the related technology about image retrieval based on content is analyzed by us. The research focused on global features such as seven HU invariant moments, edge direction histogram and eccentricity. The method for blocked image is also discussed. During the process of image matching, the extracted image features are looked as the points in vector space. The similarity of two images is measured through the closeness between two points and the similarity is calculated by Euclidean distance and the intersection distance of histogram. Then a novel method based on multi-features amalgamation is proposed, to solve the problems in retrieval method for global feature and local feature. It extracts the eccentricity, seven HU invariant moments and edge direction histogram to calculate the similarity distance of each feature of the images, then they are normalized. Contraposing the interior of global feature the weighted feature distance is adopted to form similarity measurement function for retrieval. The features of blocked images are extracted with the partitioning method based on polar coordinate. Finally by the idea of hierarchical retrieval between global feature and local feature, the results are output through global features like invariant moments etc. These results will be taken as the input of local feature match for the second-layer retrieval, which can improve the accuracy of retrieval effectively.

  19. Pancreatic neuroendocrine neoplasms: Magnetic resonance imaging features according to grade and stage

    De Robertis, Riccardo; Cingarlini, Sara; Tinazzi Martini, Paolo; Ortolani, Silvia; Butturini, Giovanni; Landoni, Luca; Regi, Paolo; Girelli, Roberto; Capelli, Paola; Gobbo, Stefano; Tortora, Giampaolo; Scarpa, Aldo; Pederzoli, Paolo; D’Onofrio, Mirko


    AIM To describe magnetic resonance (MR) imaging features of pancreatic neuroendocrine neoplasms (PanNENs) according to their grade and tumor-nodes-metastases stage by comparing them to histopathology and to determine the accuracy of MR imaging features in predicting their biological behavior. METHODS This study was approved by our institutional review board; requirement for informed patient consent was waived due to the retrospective nature of the study. Preoperative MR examinations of 55 PanNEN patients (29 men, 26 women; mean age of 57.6 years, range 21-83 years) performed between June 2013 and December 2015 were reviewed. Qualitative and quantitative features were compared between tumor grades and stages determined by histopathological analysis. RESULTS Ill defined margins were more common in G2-3 and stage III-IV PanNENs than in G1 and low-stage tumors (P < 0.001); this feature had high specificity in the identification of G2-3 and stage III-IV tumors (90.3% and 96%, 95%CI: 73.1-97.5 and 77.7-99.8). The mean apparent diffusion coefficient value was significantly lower in G2-3 and stage III-IV lesions compared to well differentiated and low-stage tumors (1.09 × 10-3 mm2/s vs 1.45 × 10-3 mm2/s and 1.10 × 10-3 mm2/s vs 1.53 × 10-3 mm2/s, P = 0.003 and 0.001). Receiving operator characteristic analysis determined optimal cut-offs of 1.21 and 1.28 × 10-3 mm2/s for the identification of G2-3 and stage III-IV tumors, with sensitivity and specificity values of 70.8/80.7% and 64.5/64% (95%CI: 48.7-86.6/60-92.7 and 45.4-80.2/42.6-81.3). CONCLUSION MR features of PanNENs vary according to their grade of differentiation and their stage at diagnosis and could predict the biological behavior of these tumors. PMID:28127201

  20. Ginkgo leaf sign: a highly predictive imaging feature of spinal meningioma.

    Yamaguchi, Satoshi; Takeda, Masaaki; Takahashi, Toshiyuki; Yamahata, Hitoshi; Mitsuhara, Takafumi; Niiro, Tadaaki; Hanakita, Junya; Hida, Kazutoshi; Arita, Kazunori; Kurisu, Kaoru


    OBJECT Spinal meningioma and schwannoma are the most common spinal intradural extramedullary tumors, and the differentiation of these 2 tumors by CT and MRI has been a matter of debate. The purpose of this article is to present a case series of spinal meningiomas showing unique imaging features: a combination of a fan-shaped spinal cord and a streak in the tumor. The authors termed the former imaging feature "ginkgo leaf sign" and evaluated its diagnostic value. METHODS The authors present 7 cases of spinal meningioma having the ginkgo leaf sign. Thirty spinal extramedullary tumors arising lateral or ventrolateral to the spinal cord were studied to evaluate the diagnostic value of the ginkgo leaf sign for spinal meningiomas. Among 30 cases, 12 tumors were spinal meningiomas and 18 tumors from the control group were all schwannomas. RESULTS Seven of the 12 spinal meningiomas were positive for the ginkgo leaf sign. The sign was not present in the control group tumors. The overall ability to use the ginkgo leaf sign to detect meningioma indicated a sensitivity of 58%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 78%. CONCLUSIONS The ginkgo leaf sign is highly specific to spinal meningiomas arising lateral or ventrolateral to the spinal cord. In the present series, the ginkgo leaf sign was perfectly predictive for spinal meningioma.

  1. Hierarchical Multi-modal Image Registration by Learning Common Feature Representations.

    Ge, Hongkun; Wu, Guorong; Wang, Li; Gao, Yaozong; Shen, Dinggang


    Mutual information (MI) has been widely used for registering images with different modalities. Since most inter-modality registration methods simply estimate deformations in a local scale, but optimizing MI from the entire image, the estimated deformations for certain structures could be dominated by the surrounding unrelated structures. Also, since there often exist multiple structures in each image, the intensity correlation between two images could be complex and highly nonlinear, which makes global MI unable to precisely guide local image deformation. To solve these issues, we propose a hierarchical inter-modality registration method by robust feature matching. Specifically, we first select a small set of key points at salient image locations to drive the entire image registration. Since the original image features computed from different modalities are often difficult for direct comparison, we propose to learn their common feature representations by projecting them from their native feature spaces to a common space, where the correlations between corresponding features are maximized. Due to the large heterogeneity between two high-dimension feature distributions, we employ Kernel CCA (Canonical Correlation Analysis) to reveal such non-linear feature mappings. Then, our registration method can take advantage of the learned common features to reliably establish correspondences for key points from different modality images by robust feature matching. As more and more key points take part in the registration, our hierarchical feature-based image registration method can efficiently estimate the deformation pathway between two inter-modality images in a global to local manner. We have applied our proposed registration method to prostate CT and MR images, as well as the infant MR brain images in the first year of life. Experimental results show that our method can achieve more accurate registration results, compared to other state-of-the-art image registration

  2. Medical Image Fusion Algorithm Based on Nonlinear Approximation of Contourlet Transform and Regional Features

    Hui Huang


    Full Text Available According to the pros and cons of contourlet transform and multimodality medical imaging, here we propose a novel image fusion algorithm that combines nonlinear approximation of contourlet transform with image regional features. The most important coefficient bands of the contourlet sparse matrix are retained by nonlinear approximation. Low-frequency and high-frequency regional features are also elaborated to fuse medical images. The results strongly suggested that the proposed algorithm could improve the visual effects of medical image fusion and image quality, image denoising, and enhancement.

  3. Importance of the texture features in a query from a spectral image database

    Kohonen, Oili; Hauta-Kasari, Markku


    A new, semantically meaningful technique for querying the images from a spectral image database is proposed. The technique is based on the use of both color- and texture features. The color features are calculated from spectral images by using the Self-Organizing Map (SOM) when methods of Gray Level Co-occurrence Matrix (GLCM) and Local Binary Pattern (LBP) are used for constructing the texture features. The importance of texture features in a querying is seen in experimental results, which are given by using a real spectral image database. Also the differences between the results gained by the use of co-occurrence matrix and LBP are introduced.

  4. MR imaging of transient synovitis: differentiation from septic arthritis

    Yang, W.J.; Im, S.A.; Lim, G.Y.; Chun, H.J.; Jung, N.Y.; Sung, M.S.; Choi, B.G. [Catholic Univ. of Korea, Seoul (Korea). Dept. of Radiology


    Transient synovitis is the most common cause of acute hip pain in children. However, MR imaging findings in transient synovitis and the role of MR imaging in differentiating transient synovitis from septic arthritis have not been fully reported. To describe the MR findings of transient synovitis and to determine whether the MR characteristics can differentiate this disease entity from septic arthritis. Clinical findings and MR images of 49 patients with transient synovitis (male/female 36/13, mean age 6.1 years) and 18 patients with septic arthritis (male/female 10/8, mean age 4.9 years) were retrospectively reviewed. MR findings of transient synovitis were symptomatic joint effusion, synovial enhancement, contralateral joint effusion, synovial thickening, and signal intensity (SI) alterations and enhancement in surrounding soft tissue. Among these, SI alterations and enhancement in bone marrow and soft tissue, contralateral joint effusion, and synovial thickening were statistically significant MR findings in differentiating transient synovitis from septic arthritis. The statistically significant MR findings in transient synovitis are contralateral (asymptomatic) joint effusions and the absence of SI abnormalities of the bone marrow. It is less common to have SI alterations and contrast enhancement of the soft tissues. The statistically significant MR findings in septic arthritis are SI alterations of the bone marrow, and SI alterations and contrast enhancement of the soft tissue. Ipsilateral effusion and synovial thickening and enhancement are present in both diseases.

  5. Emotional textile image classification based on cross-domain convolutional sparse autoencoders with feature selection

    Li, Zuhe; Fan, Yangyu; Liu, Weihua; Yu, Zeqi; Wang, Fengqin


    We aim to apply sparse autoencoder-based unsupervised feature learning to emotional semantic analysis for textile images. To tackle the problem of limited training data, we present a cross-domain feature learning scheme for emotional textile image classification using convolutional autoencoders. We further propose a correlation-analysis-based feature selection method for the weights learned by sparse autoencoders to reduce the number of features extracted from large size images. First, we randomly collect image patches on an unlabeled image dataset in the source domain and learn local features with a sparse autoencoder. We then conduct feature selection according to the correlation between different weight vectors corresponding to the autoencoder's hidden units. We finally adopt a convolutional neural network including a pooling layer to obtain global feature activations of textile images in the target domain and send these global feature vectors into logistic regression models for emotional image classification. The cross-domain unsupervised feature learning method achieves 65% to 78% average accuracy in the cross-validation experiments corresponding to eight emotional categories and performs better than conventional methods. Feature selection can reduce the computational cost of global feature extraction by about 50% while improving classification performance.

  6. Image registration algorithm using Mexican hat function-based operator and grouped feature matching strategy.

    Feng Jin

    Full Text Available Feature detection and matching are crucial for robust and reliable image registration. Although many methods have been developed, they commonly focus on only one class of image features. The methods that combine two or more classes of features are still novel and significant. In this work, methods for feature detection and matching are proposed. A Mexican hat function-based operator is used for image feature detection, including the local area detection and the feature point detection. For the local area detection, we use the Mexican hat operator for image filtering, and then the zero-crossing points are extracted and merged into the area borders. For the feature point detection, the Mexican hat operator is performed in scale space to get the key points. After the feature detection, an image registration is achieved by using the two classes of image features. The feature points are grouped according to a standardized region that contains correspondence to the local area, precise registration is achieved eventually by the grouped points. An image transformation matrix is estimated by the feature points in a region and then the best one is chosen through competition of a set of the transformation matrices. This strategy has been named the Grouped Sample Consensus (GCS. The GCS has also ability for removing the outliers effectively. The experimental results show that the proposed algorithm has high registration accuracy and small computational volume.

  7. Clinical features and magnetic resonance image analysis of 15 cases of demyelinating leukoencephalopathy induced by levamisole



    The aim of this study was to explore the diagnostic value of magnetic resonance imaging (MRI) for levamisole-induced demyelinating leukoencephalopathy. The clinical features and MRI findings of 15 patients with levamisole-induced demyelinating leukoencephalopathy were retrospectively analyzed. The abnormality rate of the patients was demonstrated to be 100% by MRI, and scattered multiple cerebral foci were observed in all of the patients. The majority of the foci were located at the centrum ovale, peri-lateral cerebral ventricles and basal ganglia, while the remainder were located in the brain stem and cerebellum, as well as in the white matter regions of the temporal, frontal, apical and occipital lobes. In addition, mottling and ring-shaped enhancements were observed. The study demonstrated that MRI effectively displays demyelinating leukoencephalopathy, and that the combination of MRI with the medical history of the patient is of significance for the early diagnosis, differentiation and treatment of demyelinating leukoencephalopathy. PMID:23935721

  8. Feature extraction for the analysis of colon status from the endoscopic images

    Krishnan Shankar M


    Full Text Available Abstract Background Extracting features from the colonoscopic images is essential for getting the features, which characterizes the properties of the colon. The features are employed in the computer-assisted diagnosis of colonoscopic images to assist the physician in detecting the colon status. Methods Endoscopic images contain rich texture and color information. Novel schemes are developed to extract new texture features from the texture spectra in the chromatic and achromatic domains, and color features for a selected region of interest from each color component histogram of the colonoscopic images. These features are reduced in size using Principal Component Analysis (PCA and are evaluated using Backpropagation Neural Network (BPNN. Results Features extracted from endoscopic images were tested to classify the colon status as either normal or abnormal. The classification results obtained show the features' capability for classifying the colon's status. The average classification accuracy, which is using hybrid of the texture and color features with PCA (τ = 1%, is 97.72%. It is higher than the average classification accuracy using only texture (96.96%, τ = 1% or color (90.52%, τ = 1% features. Conclusion In conclusion, novel methods for extracting new texture- and color-based features from the colonoscopic images to classify the colon status have been proposed. A new approach using PCA in conjunction with BPNN for evaluating the features has also been proposed. The preliminary test results support the feasibility of the proposed method.

  9. Variability of Image Features Computed from Conventional and Respiratory-Gated PET/CT Images of Lung Cancer

    Jasmine A. Oliver


    Full Text Available Radiomics is being explored for potential applications in radiation therapy. How various imaging protocols affect quantitative image features is currently a highly active area of research. To assess the variability of image features derived from conventional [three-dimensional (3D] and respiratory-gated (RG positron emission tomography (PET/computed tomography (CT images of lung cancer patients, image features were computed from 23 lung cancer patients. Both protocols for each patient were acquired during the same imaging session. PET tumor volumes were segmented using an adaptive technique which accounted for background. CT tumor volumes were delineated with a commercial segmentation tool. Using RG PET images, the tumor center of mass motion, length, and rotation were calculated. Fifty-six image features were extracted from all images consisting of shape descriptors, first-order features, and second-order texture features. Overall, 26.6% and 26.2% of total features demonstrated less than 5% difference between 3D and RG protocols for CT and PET, respectively. Between 10 RG phases in PET, 53.4% of features demonstrated percent differences less than 5%. The features with least variability for PET were sphericity, spherical disproportion, entropy (first and second order, sum entropy, information measure of correlation 2, Short Run Emphasis (SRE, Long Run Emphasis (LRE, and Run Percentage (RPC; and those for CT were minimum intensity, mean intensity, Root Mean Square (RMS, Short Run Emphasis (SRE, and RPC. Quantitative analysis using a 3D acquisition versus RG acquisition (to reduce the effects of motion provided notably different image feature values. This study suggests that the variability between 3D and RG features is mainly due to the impact of respiratory motion.

  10. Texture features on T2-weighted magnetic resonance imaging: new potential biomarkers for prostate cancer aggressiveness

    Vignati, A.; Mazzetti, S.; Giannini, V.; Russo, F.; Bollito, E.; Porpiglia, F.; Stasi, M.; Regge, D.


    To explore contrast (C) and homogeneity (H) gray-level co-occurrence matrix texture features on T2-weighted (T2w) Magnetic Resonance (MR) images and apparent diffusion coefficient (ADC) maps for predicting prostate cancer (PCa) aggressiveness, and to compare them with traditional ADC metrics for differentiating low- from intermediate/high-grade PCas. The local Ethics Committee approved this prospective study of 93 patients (median age, 65 years), who underwent 1.5 T multiparametric endorectal MR imaging before prostatectomy. Clinically significant (volume ≥0.5 ml) peripheral tumours were outlined on histological sections, contoured on T2w and ADC images, and their pathological Gleason Score (pGS) was recorded. C, H, and traditional ADC metrics (mean, median, 10th and 25th percentile) were calculated on the largest lesion slice, and correlated with the pGS through the Spearman correlation coefficient. The area under the receiver operating characteristic curve (AUC) assessed how parameters differentiate pGS = 6 from pGS ≥ 7. The dataset included 49 clinically significant PCas with a balanced distribution of pGS. The Spearman ρ and AUC values on ADC were: -0.489, 0.823 (mean) -0.522, 0.821 (median) -0.569, 0.854 (10th percentile) -0.556, 0.854 (25th percentile) -0.386, 0.871 (C); 0.533, 0.923 (H); while on T2w they were: -0.654, 0.945 (C); 0.645, 0.962 (H). AUC of H on ADC and T2w, and C on T2w were significantly higher than that of the mean ADC (p = 0.05). H and C calculated on T2w images outperform ADC parameters in correlating with pGS and differentiating low- from intermediate/high-risk PCas, supporting the role of T2w MR imaging in assessing PCa biological aggressiveness.

  11. Age Estimation Based on AAM and 2D-DCT Features of Facial Images

    Asuman Günay


    Full Text Available This paper proposes a novel age estimation method - Global and Local feAture based Age estiMation (GLAAM - relying on global and local features of facial images. Global features are obtained with Active Appearance Models (AAM. Local features are extracted with regional 2D-DCT (2- dimensional Discrete Cosine Transform of normalized facial images. GLAAM consists of the following modules: face normalization, global feature extraction with AAM, local feature extraction with 2D-DCT, dimensionality reduction by means of Principal Component Analysis (PCA and age estimation with multiple linear regression. Experiments have shown that GLAAM outperforms many methods previously applied to the FG-NET database.

  12. Forefoot pain involving the metatarsal region: differential diagnosis with MR imaging.

    Ashman, C J; Klecker, R J; Yu, J S


    Many disorders produce discomfort in the metatarsal region of the forefoot. These disorders include traumatic lesions of the soft tissues and bones (eg, turf toe, plantar plate disruption, sesamoiditis, stress fracture, stress response), Freiberg infraction, infection, arthritis, tendon disorders (eg, tendinosis, tenosynovitis, tendon rupture), nonneoplastic soft-tissue masses (eg, ganglia, bursitis, granuloma, Morton neuroma), and, less frequently, soft-tissue and bone neoplasms. Prior to the advent of magnetic resonance (MR) imaging, many of these disorders were not diagnosed noninvasively, and radiologic involvement in the evaluation of affected patients was limited. However, MR imaging has proved useful in detecting the numerous soft-tissue and early bone and joint processes that occur in this portion of the foot but are not depicted or as well characterized with other imaging modalities. Frequently, MR imaging allows a specific diagnosis based on the location, signal intensity characteristics, and morphologic features of the abnormality. Consequently, MR imaging is increasingly being used to evaluate patients with forefoot complaints. Radiologists should be familiar with the differential diagnosis and MR imaging features of disorders that can produce discomfort in this region.

  13. Fast Image Retrieval of Textile Industrial Accessory Based on Multi-Feature Fusion

    沈文忠; 杨杰


    A hierarchical retrieval scheme of the accessory image database is proposed based on textile industrial accessory contour feature and region feature. At first smallest enclosed rectangle[1] feature (degree of accessory coordination) is used to filter the image database to decouple the image search scope. After the accessory contour information and region information are extracted, the fusion multi-feature of the centroid distance Fourier descriptor and distance distribution histogram is adopted to finish image retrieval accurately. All the features above are invariable under translation, scaling and rotation. Results from the test on the image database including 1,000 accessory images demonstrate that the method is effective and practical with high accuracy and fast speed.

  14. [Research on non-rigid medical image registration algorithm based on SIFT feature extraction].

    Wang, Anna; Lu, Dan; Wang, Zhe; Fang, Zhizhen


    In allusion to non-rigid registration of medical images, the paper gives a practical feature points matching algorithm--the image registration algorithm based on the scale-invariant features transform (Scale Invariant Feature Transform, SIFT). The algorithm makes use of the image features of translation, rotation and affine transformation invariance in scale space to extract the image feature points. Bidirectional matching algorithm is chosen to establish the matching relations between the images, so the accuracy of image registrations is improved. On this basis, affine transform is chosen to complement the non-rigid registration, and normalized mutual information measure and PSO optimization algorithm are also chosen to optimize the registration process. The experimental results show that the method can achieve better registration results than the method based on mutual information.

  15. Feature Extraction with Ordered Mean Values for Content Based Image Classification

    Sudeep Thepade


    Full Text Available Categorization of images into meaningful classes by efficient extraction of feature vectors from image datasets has been dependent on feature selection techniques. Traditionally, feature vector extraction has been carried out using different methods of image binarization done with selection of global, local, or mean threshold. This paper has proposed a novel technique for feature extraction based on ordered mean values. The proposed technique was combined with feature extraction using discrete sine transform (DST for better classification results using multitechnique fusion. The novel methodology was compared to the traditional techniques used for feature extraction for content based image classification. Three benchmark datasets, namely, Wang dataset, Oliva and Torralba (OT-Scene dataset, and Caltech dataset, were used for evaluation purpose. Performance measure after evaluation has evidently revealed the superiority of the proposed fusion technique with ordered mean values and discrete sine transform over the popular approaches of single view feature extraction methodologies for classification.

  16. Multimodal imaging in the differential diagnosis of soft tissue calcinosis

    G. Garlaschi


    Full Text Available Soft tissue calcinosis is a common radiographic finding, which may be related to different types of pathological processes. Multimodality imaging, combined with analysis of clinical and laboratory data, plays an important role for the differential diagnosis of these conditions. Conventional radiography is considered the first line approach to soft tissue calcinosis; CT and MRI may provide further information to better characterize calcified deposits. Imaging may help to distinguish metabolic calcification, such as primary tumoral calcinosis and the secondary one (associated with acquired disorders of calcium or phosphate regulation, from dystrophic calcification, which is associated to normal blood values of phosphate. The sedimentation sign typical of tumoral calcinosis has been demonstrated by plain film radiography, CT, MRI, and, more recently, by ultrasonography. Other types of soft tissue calcinosis may have a degenerative, metaplastic or neoplastic origin, and their characterization strongly relies on multimodality imaging.

  17. In vivo hyperspectral imaging and differentiation of skin cancer

    Zherdeva, Larisa A.; Bratchenko, Ivan A.; Myakinin, Oleg O.; Moryatov, Alexander A.; Kozlov, Sergey V.; Zakharov, Valery P.


    Results of hyperspectral imaging analysis for in vivo visualization of skin neoplasms are presented. 16 melanomas, 19 basal cell carcinomas and 10 benign tumors with different stages of neoplasm growth were tested. The HSI system provide skin tissue images with 5 nm spectral resolution in the range of 450-750 nm with automatic stabilization of each frame compensating displacement of the scanning area due to spontaneous macro-movements of the patient. The integrated optical densities in 530-600 and 600-670 nm ranges are used for real-time hemoglobin and melanin distribution imaging in skin tissue. It was shown that the total accuracy of skin cancer identification exceeds 90% and 70% for differentiation of melanomas from BCC and begihn tumors. It was demonstrated the possibility for HSI classification of melanomas of different stages.

  18. Automated oral cancer identification using histopathological images: a hybrid feature extraction paradigm.

    Krishnan, M Muthu Rama; Venkatraghavan, Vikram; Acharya, U Rajendra; Pal, Mousumi; Paul, Ranjan Rashmi; Min, Lim Choo; Ray, Ajoy Kumar; Chatterjee, Jyotirmoy; Chakraborty, Chandan


    Oral cancer (OC) is the sixth most common cancer in the world. In India it is the most common malignant neoplasm. Histopathological images have widely been used in the differential diagnosis of normal, oral precancerous (oral sub-mucous fibrosis (OSF)) and cancer lesions. However, this technique is limited by subjective interpretations and less accurate diagnosis. The objective of this work is to improve the classification accuracy based on textural features in the development of a computer assisted screening of OSF. The approach introduced here is to grade the histopathological tissue sections into normal, OSF without Dysplasia (OSFWD) and OSF with Dysplasia (OSFD), which would help the oral onco-pathologists to screen the subjects rapidly. The biopsy sections are stained with H&E. The optical density of the pixels in the light microscopic images is recorded and represented as matrix quantized as integers from 0 to 255 for each fundamental color (Red, Green, Blue), resulting in a M×N×3 matrix of integers. Depending on either normal or OSF condition, the image has various granular structures which are self similar patterns at different scales termed "texture". We have extracted these textural changes using Higher Order Spectra (HOS), Local Binary Pattern (LBP), and Laws Texture Energy (LTE) from the histopathological images (normal, OSFWD and OSFD). These feature vectors were fed to five different classifiers: Decision Tree (DT), Sugeno Fuzzy, Gaussian Mixture Model (GMM), K-Nearest Neighbor (K-NN), Radial Basis Probabilistic Neural Network (RBPNN) to select the best classifier. Our results show that combination of texture and HOS features coupled with Fuzzy classifier resulted in 95.7% accuracy, sensitivity and specificity of 94.5% and 98.8% respectively. Finally, we have proposed a novel integrated index called Oral Malignancy Index (OMI) using the HOS, LBP, LTE features, to diagnose benign or malignant tissues using just one number. We hope that this OMI can

  19. Discriminative feature representation: an effective postprocessing solution to low dose CT imaging

    Chen, Yang; Liu, Jin; Hu, Yining; Yang, Jian; Shi, Luyao; Shu, Huazhong; Gui, Zhiguo; Coatrieux, Gouenou; Luo, Limin


    This paper proposes a concise and effective approach termed discriminative feature representation (DFR) for low dose computerized tomography (LDCT) image processing, which is currently a challenging problem in medical imaging field. This DFR method assumes LDCT images as the superposition of desirable high dose CT (HDCT) 3D features and undesirable noise-artifact 3D features (the combined term of noise and artifact features induced by low dose scan protocols), and the decomposed HDCT features are used to provide the processed LDCT images with higher quality. The target HDCT features are solved via the DFR algorithm using a featured dictionary composed by atoms representing HDCT features and noise-artifact features. In this study, the featured dictionary is efficiently built using physical phantom images collected from the same CT scanner as the target clinical LDCT images to process. The proposed DFR method also has good robustness in parameter setting for different CT scanner types. This DFR method can be directly applied to process DICOM formatted LDCT images, and has good applicability to current CT systems. Comparative experiments with abdomen LDCT data validate the good performance of the proposed approach. This research was supported by National Natural Science Foundation under grants (81370040, 81530060), the Fundamental Research Funds for the Central Universities, and the Qing Lan Project in Jiangsu Province.

  20. Nuclear Molecular and Theranostic Imaging for Differentiated Thyroid Cancer

    Arif Sheikh


    Full Text Available Traditional nuclear medicine is rapidly being transformed by the evolving concepts in molecular imaging and theranostics. The utility of new approaches in differentiated thyroid cancer (DTC diagnostics and therapy has not been fully appreciated. The clinical information, relevant to disease management and patient care, obtained by scintigraphy is still being underestimated. There has been a trend towards moving away from the use of radioactive iodine (RAI imaging in the management of the disease. This paradigm shift is supported by the 2015 American Thyroid Association Guidelines (1. A more systematic and comprehensive understanding of disease pathophysiology and imaging methodologies is needed for optimal utilization of different imaging modalities in the management of DTC. There have been significant developments in radiotracer and imaging technology, clinically proven to contribute to the understanding of tumor biology and the clinical assessment of patients with DTC. The research and development in the field continues to evolve, with expected emergence of many novel diagnostic and therapeutic techniques. The role for nuclear imaging applications will continue to evolve and be reconfigured in the changing paradigm. This article aims to review the clinical uses and controversies surrounding the use of scintigraphy, and the information it can provide in assisting in the management and treatment of DTC.

  1. The Sensitivity of Hybrid Differential Stereoscopy for Spectral Imaging

    DeForest, Craig E


    Stereoscopic spectral imaging is an observing technique that affords rapid acquisition of limited spectral information over an entire image plane simultaneously. Light from a telescope is dispersed into multiple spectral orders, which are imaged separately, and two or more of the dispersed images are combined using an analogy between the (x,y,\\lambda) spectral data space and conventional (x,y,z) three-space. Because no photons are deliberately destroyed during image acquisition, the technique is much more photon-efficient in some observing regimes than existing techniques such as scanned-filtergraph or scanned-slit spectral imaging. Hybrid differential stereoscopy, which uses a combination of conventional cross-correlation stereoscopy and linear approximation theory to extract the central wavelength of a spectral line, has been used to produce solar Stokes-V (line-of-sight) magnetograms in the 617.34 nm Fe I line, and more sophisticated inversion techniques are currently being used to derive Doppler and line ...

  2. [Adrenal tumors: principles of imaging and differential diagnostics].

    Degenhart, C


    Adrenal masses are very common and are usually detected incidentally. Less frequently, imaging is performed for the localization of the underlying lesion in the case of endocrine disease. The differentiation between adenomas and non-adenomas is fundamental. Adenomas show a low density on unenhanced computed tomography (CT) and a rapid washout of contrast agents. In magnetic resonance imaging (MRI) adenomas are characterized by a low signal in opposed phase imaging as compared to in phase imaging. According to the literature a density of less than 10 HU in an adrenal mass has a specificity of 98% and a sensitivity of 71% for the presence of an adenoma and MRI is slightly more sensitive. Some adrenal lesions, e.g. cysts or myelolipomas can be diagnosed with high accuracy due to pathognomonic findings. In the majority of cases the synopsis of imaging along with clinical and laboratory findings is necessary for a reliable diagnosis. For the evaluation of an adrenal mass the CT examination should begin with an unenhanced scan, if necessary followed by a washout examination. In the case of MRI in phase and opposed phase imaging are essential components of the examination.

  3. Differentiation between ovarian fibroma and subserosal leiomyoma by MR imaging

    Choi, Sang Yeol; Lee, Jun Woo; Kim, Chang Won; Kim, Yong Woo; Lee, Suck Hong [College of Medicine, Pusan National University, Pusan (Korea, Republic of)


    To evaluate the findings and differential points of ovarian fibroma and subserosal leiomyoma, as seen on MR images. The MRimaging findings of 31 surgically confirmed cases of ovarian fibroma (n=3D6) and subserosal leiomyoma (n=3D25; 28) lesions were evaluated. Multiplanar T1-T2-weighted and postcontrast T1-weighted images were obtained using a 1.5T MR unit, and histologic examination was also performed. The MR findings were analyzed in terms of signal intensity, the presence and definition of margin, the histologic finding of hyperintense lesion on T2-weighted images, the presence of the bridging vessel sign, degree of enhancement, and the presence of ipsilateral ovary and ascites. Both fibromas and leiomyomas showed hypo- or isointensity compared with uterine myometrium on T1-weighted images and compared with skeletal muscle on T2-weighted images. The latter revealed intratumoral hyperintense lesions in most cases of ovarian fibroma and subserosal leiomyoma. Three of four ovarian fibromas had a well defined margin after cystic change, but in 24 of 26 subserosal leiomyomas the margin was ill defined. The 'bridging vessel sign' was visible only in subserosal leiomyomas (22/28), and in all cases the enhancement of ovarian fibromas were less than that of myomtetrium. Subserosal leiomyomas (12/28), seen on enhancement as isointense or hyperintense to myometrium, showed a greater degree of enhancement than ovarian fibromas (0/6). Ipsilateral ovary was rarely seen in ovarian fibromas (1/6), but commonly seen in subserosal leiomyomas (20/250. Ascites was present in one case of ovarian fibroma. A defined margin of an intratumoral hyperintense lesion, as seen on T2-weighted images, and the presence or absence of the 'bridging vessel sign' and ipsilateral ovary are useful signs when differentiating between ovarian fibromas and subserosal leiomyomas. (author)

  4. Multiple Myeloma: A Review of Imaging Features and Radiological Techniques

    C. F. Healy


    Full Text Available The recently updated Durie/Salmon PLUS staging system published in 2006 highlights the many advances that have been made in the imaging of multiple myeloma, a common malignancy of plasma cells. In this article, we shall focus primarily on the more sensitive and specific whole-body imaging techniques, including whole-body computed tomography, whole-body magnetic resonance imaging, and positron emission computed tomography. We shall also discuss new and emerging imaging techniques and future developments in the radiological assessment of multiple myeloma.

  5. Multiple myeloma: a review of imaging features and radiological techniques.

    Healy, C F; Murray, J G; Eustace, S J; Madewell, J; O'Gorman, P J; O'Sullivan, P


    The recently updated Durie/Salmon PLUS staging system published in 2006 highlights the many advances that have been made in the imaging of multiple myeloma, a common malignancy of plasma cells. In this article, we shall focus primarily on the more sensitive and specific whole-body imaging techniques, including whole-body computed tomography, whole-body magnetic resonance imaging, and positron emission computed tomography. We shall also discuss new and emerging imaging techniques and future developments in the radiological assessment of multiple myeloma.

  6. Gilbert’s syndrome: clinical features, diagnostics, differential diagnosis and treatment (part 2

    T.V. Sorokman


    discoloration of the skin (“teinte bilieuse”, especially on the face, hands, and feet without a distinct scleral icterus. Sometimes the development of repeatedly intermittent episodes of jaundice with high bilirubinemia (indirect bilirubin without the evidence of hemolysis (differential diagnostic feature is observed. 2. A tendency to development of pigmented and vascular nevi and xanthelasma of the eyelids, and hyperpigmentation around the eyes; to bradycardia, hypothermia, migraine, postural, intermittent albuminuria or to alimentary glycosuria. 3. An increased tendency to pigmentation under the influence of light, heat, and also chemical and mechanical stimuli. 4. A neuromuscular hyperexcitability. 5. Increased sensitivity to cold. 6. Dyspeptic complaints (pain, nausea, abdominal bloa­ting, diarrhea or constipation. 7. No signs of increased hemolysis (differential diagnostic feature with increasing content in, bilirubin (differential diagnostic feature. 8. The majority of patients have normal liver function tests (differential diagnostic feature also normal bromsulphalein test is also normal (differential diagnostic feature. 9. The biochemical abnormality is not detected by histological methods (differential diagnostic feature .10. Frequently, a family disease of the liver is observed. The differential diagnosis of GS is conducted with all types of hyperbilirubinemias, hemolytic anemias, congenital hepatic cirrhosis, hepatitis, cholecystopathy, atresia of biliary ducts or the small intestine. Medications are used only in severe hyperbilirubinemias and as concomitant therapy in the presence of symptoms of vitamin deficiencies, violations of a motor-evacuation function of the upper digestive tract in the clinical picture and to prevent complications (cholelithiasis.

  7. Change Detection in Uav Video Mosaics Combining a Feature Based Approach and Extended Image Differencing

    Saur, Günter; Krüger, Wolfgang


    Change detection is an important task when using unmanned aerial vehicles (UAV) for video surveillance. We address changes of short time scale using observations in time distances of a few hours. Each observation (previous and current) is a short video sequence acquired by UAV in near-Nadir view. Relevant changes are, e.g., recently parked or moved vehicles. Examples for non-relevant changes are parallaxes caused by 3D structures of the scene, shadow and illumination changes, and compression or transmission artifacts. In this paper we present (1) a new feature based approach to change detection, (2) a combination with extended image differencing (Saur et al., 2014), and (3) the application to video sequences using temporal filtering. In the feature based approach, information about local image features, e.g., corners, is extracted in both images. The label "new object" is generated at image points, where features occur in the current image and no or weaker features are present in the previous image. The label "vanished object" corresponds to missing or weaker features in the current image and present features in the previous image. This leads to two "directed" change masks and differs from image differencing where only one "undirected" change mask is extracted which combines both label types to the single label "changed object". The combination of both algorithms is performed by merging the change masks of both approaches. A color mask showing the different contributions is used for visual inspection by a human image interpreter.

  8. Image mosaicking based on feature points using color-invariant values

    Lee, Dong-Chang; Kwon, Oh-Seol; Ko, Kyung-Woo; Lee, Ho-Young; Ha, Yeong-Ho


    In the field of computer vision, image mosaicking is achieved using image features, such as textures, colors, and shapes between corresponding images, or local descriptors representing neighborhoods of feature points extracted from corresponding images. However, image mosaicking based on feature points has attracted more recent attention due to the simplicity of the geometric transformation, regardless of distortion and differences in intensity generated by camera motion in consecutive images. Yet, since most feature-point matching algorithms extract feature points using gray values, identifying corresponding points becomes difficult in the case of changing illumination and images with a similar intensity. Accordingly, to solve these problems, this paper proposes a method of image mosaicking based on feature points using color information of images. Essentially, the digital values acquired from a real digital color camera are converted to values of a virtual camera with distinct narrow bands. Values based on the surface reflectance and invariant to the chromaticity of various illuminations are then derived from the virtual camera values and defined as color-invariant values invariant to changing illuminations. The validity of these color-invariant values is verified in a test using a Macbeth Color-Checker under simulated illuminations. The test also compares the proposed method using the color-invariant values with the conventional SIFT algorithm. The accuracy of the matching between the feature points extracted using the proposed method is increased, while image mosaicking using color information is also achieved.

  9. Comparing differential tissue harmonic imaging with tissue harmonic and fundamental gray scale imaging of the liver.

    Chiou, See-Ying; Forsberg, Flemming; Fox, Traci B; Needleman, Laurence


    The purpose of this study was to compare fundamental gray scale sonography, tissue harmonic imaging (THI), and differential tissue harmonic imaging (DTHI) for depicting normal and abnormal livers. The in vitro lateral resolution of DTHI, THI, and sonography was assessed in a phantom. Sagittal and transverse images of right and left hepatic lobes of 5 volunteers and 20 patients and images of 27 liver lesions were also acquired. Three independent blinded readers scored all randomized images for noise, detail resolution, image quality, and margin (for lesions) on a 7-point scale. Next, images from the same location obtained with all 3 modes were compared blindly side by side and rated by all readers. Contrast-to-noise ratios were calculated for the lesions, and the depth of penetration (centimeters) was determined for all images. In vitro, the lateral resolution of DTHI was significantly better than fundamental sonography (P = .009) and showed a trend toward significance versus THI (P = .06). In the far field, DTHI performed better than both modes (P images were scored, and for all parameters, DTHI and THI did better than sonography (P tissue harmonic imaging scored significantly higher than THI with regard to detail resolution and image quality (P Tissue harmonic imaging and DTHI do better than fundamental sonography for hepatic imaging, with DTHI being rated the best of the 3 techniques.

  10. An Adequate Approach to Image Retrieval Based on Local Level Feature Extraction

    Sumaira Muhammad Hayat Khan


    Full Text Available Image retrieval based on text annotation has become obsolete and is no longer interesting for scientists because of its high time complexity and low precision in results. Alternatively, increase in the amount of digital images has generated an excessive need for an accurate and efficient retrieval system. This paper proposes content based image retrieval technique at a local level incorporating all the rudimentary features. Image undergoes the segmentation process initially and each segment is then directed to the feature extraction process. The proposed technique is also based on image?s content which primarily includes texture, shape and color. Besides these three basic features, FD (Fourier Descriptors and edge histogram descriptors are also calculated to enhance the feature extraction process by taking hold of information at the boundary. Performance of the proposed method is found to be quite adequate when compared with the results from one of the best local level CBIR (Content Based Image Retrieval techniques.

  11. Application of Fisher Score and mRMR Techniques for Feature Selection in Compressed Medical Images

    Vamsidhar Enireddy


    Full Text Available In nowadays there is a large increase in the digital medical images and different medical imaging equipments are available for diagnoses, medical professionals are increasingly relying on computer aided techniques for both indexing these images and retrieving similar images from large repositories. To develop systems which are computationally less intensive without compromising on the accuracy from the high dimensional feature space is always challenging. In this paper an investigation is made on the retrieval of compressed medical images. Images are compressed using the visually lossless compression technique. Shape and texture features are extracted and best features are selected using the fisher technique and mRMR. Using these selected features RNN with BPTT was utilized for classification of the compressed images.

  12. Dynamic contrast-enhanced MR imaging features of the normal central zone of the prostate.

    Hansford, Barry G; Karademir, Ibrahim; Peng, Yahui; Jiang, Yulei; Karczmar, Gregory; Thomas, Stephen; Yousuf, Ambereen; Antic, Tatjana; Eggener, Scott; Oto, Aytekin


    Evaluate qualitative dynamic contrast-enhanced magnetic resonance imaging (MRI) characteristics of normal central zone based on recently described central zone MRI features. Institutional review board-approved, Health Insurance Portability and Accountability Act compliant study, 59 patients with prostate cancer, histopathology proven to not involve central zone or prostate base, underwent endorectal MRI before prostatectomy. Two readers independently reviewed T2-weighted images and apparent diffusion coefficient (ADC) maps identifying normal central zone based on low signal intensity and location. Next, two readers drew bilateral central zone regions of interest on dynamic contrast-enhanced magnetic resonance images in consensus and independently recorded enhancement curve types as type 1 (progressive), type 2 (plateau), and type 3 (wash-out). Identification rates of normal central zone and enhancement curve type were recorded and compared for each reviewer. The institutional review board waiver was approved and granted 05/2010. Central zone identified in 92%-93% of patients on T2-weighted images and 78%-88% on ADC maps without significant difference between identification rates (P = .63 and P = .15 and inter-reader agreement (κ) is 0.64 and 0.29, for T2-weighted images and ADC maps, respectively). All central zones were rated either curve type 1 or curve type 2 by both radiologists. No statistically significant difference between the two radiologists (P = .19) and inter-reader agreement was κ = 0.37. Normal central zone demonstrates either type 1 (progressive) or type 2 (plateau) enhancement curves on dynamic contrast-enhanced MRI that can be potentially useful to differentiate central zone from prostate cancer that classically demonstrates a type 3 (wash-out) enhancement curve. Copyright © 2014 AUR. Published by Elsevier Inc. All rights reserved.

  13. A Novel Technique for Shape Feature Extraction Using Content Based Image Retrieval

    Dhanoa Jaspreet Singh


    Full Text Available With the advent of technology and multimedia information, digital images are increasing very quickly. Various techniques are being developed to retrieve/search digital information or data contained in the image. Traditional Text Based Image Retrieval System is not plentiful. Since it is time consuming as it require manual image annotation. Also, the image annotation differs with different peoples. An alternate to this is Content Based Image Retrieval (CBIR system. It retrieves/search for image using its contents rather the text, keywords etc. A lot of exploration has been compassed in the range of Content Based Image Retrieval (CBIR with various feature extraction techniques. Shape is a significant image feature as it reflects the human perception. Moreover, Shape is quite simple to use by the user to define object in an image as compared to other features such as Color, texture etc. Over and above, if applied alone, no descriptor will give fruitful results. Further, by combining it with an improved classifier, one can use the positive features of both the descriptor and classifier. So, a tryout will be made to establish an algorithm for accurate feature (Shape extraction in Content Based Image Retrieval (CBIR. The main objectives of this project are: (a To propose an algorithm for shape feature extraction using CBIR, (b To evaluate the performance of proposed algorithm and (c To compare the proposed algorithm with state of art techniques.

  14. Multiphase contrast-enhanced magnetic resonance imaging features of Bacillus Calmette-Guerin-induced granulomatous prostatitis in five patients

    Kawada, Hiroshi; Kanematsu, Masayuki; Goshima, Satoshi; Kondo, Hiroshi; Watanabe, Haruo; Noda, Yoshifumi; Tanahashi, Yukichi; Kawai, Nobuyuki; Hoshi, Hiroaki [Gifu University Hospital, Gifu (Japan)


    To evaluate the multiphase contrast-enhanced magnetic resonance (MR) imaging features of Bacillus Calmette-Guerin (BCG)-induced granulomatous prostatitis (GP). Magnetic resonance images obtained from five patients with histopathologically proven BCG-induced GP were retrospectively analyzed for tumor location, size, signal intensity on T2-weighted images (T2WI) and diffusion-weighted images (DWI), apparent diffusion coefficient (ADC) value, and appearance on gadolinium-enhanced multiphase images. MR imaging findings were compared with histopathological findings. Bacillus Calmette-Guerin-induced GP (size range, 9-40 mm; mean, 21.2 mm) were identified in the peripheral zone in all patients. The T2WI showed lower signal intensity compared with the normal peripheral zone. The DWIs demonstrated high signal intensity and low ADC values (range, 0.44-0.68 x 10(-3) mm2/sec; mean, 0.56 x 10(-3) mm2/sec), which corresponded to GP. Gadolinium-enhanced multiphase MR imaging performed in five patients showed early and prolonged ring enhancement in all cases of GP. Granulomatous tissues with central caseation necrosis were identified histologically, which corresponded to ring enhancement and a central low intensity area on gadolinium-enhanced MR imaging. The findings on T2WI, DWI, and gadolinium-enhanced images became gradually obscured with time. Bacillus Calmette-Guerin-induced GP demonstrates early and prolonged ring enhancement on gadolinium-enhanced MR imaging which might be a key finding to differentiate it from prostate cancer.

  15. Image feature meaning for automatic key-frame extraction

    Di Lecce, Vincenzo; Guerriero, Andrea


    Video abstraction and summarization, being request in several applications, has address a number of researches to automatic video analysis techniques. The processes for automatic video analysis are based on the recognition of short sequences of contiguous frames that describe the same scene, shots, and key frames representing the salient content of the shot. Since effective shot boundary detection techniques exist in the literature, in this paper we will focus our attention on key frames extraction techniques to identify the low level visual features of the frames that better represent the shot content. To evaluate the features performance, key frame automatically extracted using these features, are compared to human operator video annotations.

  16. Essential features of Chiari II malformation in MR imaging : an interobserver reliability study-part 1

    Geerdink, Niels; van der Vliet, Ton; Rotteveel, Jan J.; Feuth, Ton; Roeleveld, Nel; Mullaart, Reinier A.


    Brain MR imaging is essential in the assessment of Chiari II malformation in clinical and research settings concerning spina bifida. However, the interpretation of morphological features of the malformation on MR images may not always be straightforward. In an attempt to select those features that u

  17. Essential features of Chiari II malformation in MR imaging: an interobserver reliability study--part 1.

    Geerdink, N.; Vliet, T. van der; Rotteveel, J.J.; Feuth, T.; Roeleveld, N.; Mullaart, R.A.


    PURPOSE: Brain MR imaging is essential in the assessment of Chiari II malformation in clinical and research settings concerning spina bifida. However, the interpretation of morphological features of the malformation on MR images may not always be straightforward. In an attempt to select those featur


    Yogapriya Jaganathan


    Full Text Available For the past few years, massive upgradation is obtained in the pasture of Content Based Medical Image Retrieval (CBMIR for effective utilization of medical images based on visual feature analysis for the purpose of diagnosis and educational research. The existing medical image retrieval systems are still not optimal to solve the feature dimensionality reduction problem which increases the computational complexity and decreases the speed of a retrieval process. The proposed CBMIR is used a hybrid approach based on Feature Extraction, Optimization of Feature Vectors, Classification of Features and Similarity Measurements. This type of CBMIR is called Feature Optimized Classification Similarity (FOCS framework. The selected features are Textures using Gray level Co-occurrence Matrix Features (GLCM and Tamura Features (TF in which extracted features are formed as feature vector database. The Fuzzy based Particle Swarm Optimization (FPSO technique is used to reduce the feature vector dimensionality and classification is performed using Fuzzy based Relevance Vector Machine (FRVM to form groups of relevant image features that provide a natural way to classify dimensionally reduced feature vectors of images. The Euclidean Distance (ED is used as similarity measurement to measure the significance between the query image and the target images. This FOCS approach can get the query from the user and has retrieved the needed images from the databases. The retrieval algorithm performances are estimated in terms of precision and recall. This FOCS framework comprises several benefits when compared to existing CBMIR. GLCM and TF are used to extract texture features and form a feature vector database. Fuzzy-PSO is used to reduce the feature vector dimensionality issues while selecting the important features in the feature vector database in which computational complexity is decreased. Fuzzy based RVM is used for feature classification in which it increases the

  19. Identification of the Causative Disease of Intermittent Claudication through Walking Motion Analysis: Feature Analysis and Differentiation

    Tetsuyou Watanabe


    Full Text Available Intermittent claudication is a walking symptom. Patients with intermittent claudication experience lower limb pain after walking for a short time. However, rest relieves the pain and allows the patient to walk again. Unfortunately, this symptom predominantly arises from not 1 but 2 different diseases: LSS (lumber spinal canal stenosis and PAD (peripheral arterial disease. Patients with LSS can be subdivided by the affected vertebra into 2 main groups: L4 and L5. It is clinically very important to determine whether patients with intermittent claudication suffer from PAD, L4, or L5. This paper presents a novel SVM- (support vector machine- based methodology for such discrimination/differentiation using minimally required data, simple walking motion data in the sagittal plane. We constructed a simple walking measurement system that is easy to set up and calibrate and suitable for use by nonspecialists in small spaces. We analyzed the obtained gait patterns and derived input parameters for SVM that are also visually detectable and medically meaningful/consistent differentiation features. We present a differentiation methodology utilizing an SVM classifier. Leave-one-out cross-validation of differentiation/classification by this method yielded a total accuracy of 83%.

  20. Integrating Color and Spatial Feature for Content-Based Image Retrieval


    In this paper, we present a novel and efficient scheme for extracting, indexing and retrieving color images. Our motivation was to reduce the space overhead of partition-based approaches taking advantage of the fact that only a relatively low number of distinct values of a particular visual feature is present in most images. To extract color feature and build indices into our image database we take into consideration factors such as human color perception and perceptual range, and the image is partitioned into a set of regions by using a simple classifying scheme. The compact color feature vector and the spatial color histogram, which are extracted from the seqmented image region, are used for representing the color and spatial information in the image. We have also developed the region-based distance measures to compare the similarity of two images. Extensive tests on a large image collection were conducted to demonstrate the effectiveness of the proposed approach.

  1. Fuzzy zoning for feature matching technique in 3D reconstruction of nasal endoscopic images.

    Rattanalappaiboon, Surapong; Bhongmakapat, Thongchai; Ritthipravat, Panrasee


    3D reconstruction from nasal endoscopic images greatly supports an otolaryngologist in examining nasal passages, mucosa, polyps, sinuses, and nasopharyx. In general, structure from motion is a popular technique. It consists of four main steps; (1) camera calibration, (2) feature extraction, (3) feature matching, and (4) 3D reconstruction. Scale Invariant Feature Transform (SIFT) algorithm is normally used for both feature extraction and feature matching. However, SIFT algorithm relatively consumes computational time particularly in the feature matching process because each feature in an image of interest is compared with all features in the subsequent image in order to find the best matched pair. A fuzzy zoning approach is developed for confining feature matching area. Matching between two corresponding features from different images can be efficiently performed. With this approach, it can greatly reduce the matching time. The proposed technique is tested with endoscopic images created from phantoms and compared with the original SIFT technique in terms of the matching time and average errors of the reconstructed models. Finally, original SIFT and the proposed fuzzy-based technique are applied to 3D model reconstruction of real nasal cavity based on images taken from a rigid nasal endoscope. The results showed that the fuzzy-based approach was significantly faster than traditional SIFT technique and provided similar quality of the 3D models. It could be used for creating a nasal cavity taken by a rigid nasal endoscope.

  2. Features of the differential diagnosis of persons with gender identity disorders

    Z.D. Novikova


    Full Text Available We presented a study of the features of gender identity in people undergoing gender, psychological and psychiatric examination to address the issue of gender reassignment. We analyze the specifics of gender identity, levels of masculinity and femininity, the similarities and differentiation within four nosological groups, which include persons with gender identity disorders (GID with transsexualism, personality disorders, diseases of the schizophrenia spectrum, and with organic mental disorders. We address the question of the differential diagnosis in the process of psychological screening of people with transsexualism and other types of GID. The analytical description of the four algorithms and their comparison are psychologically specific, qualitative research, almost impossible using statistical method of data processing. The data presented may be useful to specialists involved in the study of persons with gender identity disorders

  3. Imaging of the dopaminergic system in differential diagnosis of dementia

    Tatsch, Klaus [University of Munich Hospital - Campus Grosshadern, Department of Nuclear Medicine, Munich (Germany)


    Neurodegenerative dementia is an increasingly common disorder with Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) accounting for most cases. Due to the overlap in clinical symptoms, their differential diagnosis may be challenging. As clinical classification is not completely satisfying, there is a need to improve the diagnostic accuracy by complementary methods such as functional single-photon emission computed tomography (SPECT) and positron emission tomography (PET) imaging. The latter may be helpful to address one distinct biological difference between DLB and AD, the severe nigrostriatal degeneration which occurs in DLB, but not to any significant extent in AD. Based on this principle, autoradiographic studies targeting presynaptic dopaminergic functions have consistently demonstrated the ability to distinguish DLB from AD in postmortem series. At the same time, several single-site and one multicentre study have independently confirmed - no matter what technique was used (SPECT or PET) and which presynaptic function was addressed (dopamine turnover, dopamine transporter, vesicular monoamine transporter) - significantly compromised scan results in DLB subjects, whereas AD patients maintained almost normal findings. Even more important, in vivo findings of presynaptic dopaminergic imaging correlated well with neuropathological findings at autopsy, suggesting a remarkable sensitivity of 88% and a specificity of 100% for the imaging procedure to distinguish between DLB and AD. Taken together, imaging of presynaptic dopaminergic terminal functions with SPECT and PET has currently the greatest evidence base to support its use, and therefore, may be highly recommended to help in the discrimination between DLB and AD. Compared to presynaptic functions, corresponding data targeting postsynaptic dopamine receptors are comparatively rare, less conclusive and suggest a very limited role for this purpose. This review discusses the findings of studies

  4. Differentiating fatty and non-fatty tissue using photoacoustic imaging

    Pan, Leo; Rohling, Robert; Abolmaesumi, Purang; Salcudean, Septimiu; Tang, Shuo


    In this paper, we demonstrate a temporal-domain intensity-based photoacoustic imaging method that can differentiate between fatty and non-fatty tissues. PA pressure intensity is partly dependent on the tissue's speed of sound, which increases as temperature increases in non-fatty tissue and decreases in fatty tissue. Therefore, by introducing a temperature change in the tissue and subsequently monitoring the change of the PA intensity, it is possible to distinguish between the two types of tissue. A commercial ultrasound system with a 128-element 5-14 MHz linear array transducer and a tunable ND:YAG laser were used to produce PA images. Ex-vivo bovine fat and porcine liver tissues were precooled to below 10°C and then warmed to room-temperature over ~1 hour period. A thermocouple monitored the temperature rise while PA images were acquired at 0.5°C intervals. The averaged intensity of the illuminated tissue region at each temperature interval was plotted and linearly fitted. Liver samples showed a mean increase of 2.82 %/°C, whereas bovine fat had a mean decrease of 6.24 %/°C. These results demonstrate that this method has the potential to perform tissue differentiation in the temporal-domain.

  5. Face image analysis using a multiple features fitting strategy

    Romdhani, Sami


    The main contribution of this thesis is a novel algorithm for fitting a Three-Dimensional Morphable Model of faces to a 2D input image. This fitting algorithm enables the estimation of the 3D shape, the texture, the 3D pose and the light direction from a single input image. Generally, the algorithms tackling the problem of 3D shape estimation from image data use only the pixels intensity as input to drive the estimation process. This was previously achieved using either a simple model, such as ...

  6. Kernel Density Feature Points Estimator for Content-Based Image Retrieval

    Zuva, Tranos; Ojo, Sunday O; Ngwira, Seleman M


    Research is taking place to find effective algorithms for content-based image representation and description. There is a substantial amount of algorithms available that use visual features (color, shape, texture). Shape feature has attracted much attention from researchers that there are many shape representation and description algorithms in literature. These shape image representation and description algorithms are usually not application independent or robust, making them undesirable for generic shape description. This paper presents an object shape representation using Kernel Density Feature Points Estimator (KDFPE). In this method, the density of feature points within defined rings around the centroid of the image is obtained. The KDFPE is then applied to the vector of the image. KDFPE is invariant to translation, scale and rotation. This method of image representation shows improved retrieval rate when compared to Density Histogram Feature Points (DHFP) method. Analytic analysis is done to justify our m...

  7. Modeling multiple visual words assignment for bag-of-features based medical image retrieval

    Wang, Jim Jing-Yan


    In this paper, we investigate the bag-of-features based medical image retrieval methods, which represent an image as a collection of local features, such as image patch and key points with SIFT descriptor. To improve the bag-of-features method, we first model the assignment of local descriptor as contribution functions, and then propose a new multiple assignment strategy. By assuming the local feature can be reconstructed by its neighboring visual words in vocabulary, we solve the reconstruction weights as a QP problem and then use the solved weights as contribution functions, which results in a new assignment method called the QP assignment. We carry our experiments on ImageCLEFmed datasets. Experiments\\' results show that our proposed method exceeds the performances of traditional solutions and works well for the bag-of-features based medical image retrieval tasks.

  8. 3D Elastic Registration of Ultrasound Images Based on Skeleton Feature

    LI Dan-dan; LIU Zhi-Yan; SHEN Yi


    In order to eliminate displacement and elastic deformation between images of adjacent frames in course of 3D ultrasonic image reconstruction, elastic registration based on skeleton feature was adopt in this paper. A new automatically skeleton tracking extract algorithm is presented, which can extract connected skeleton to express figure feature. Feature points of connected skeleton are extracted automatically by accounting topical curvature extreme points several times. Initial registration is processed according to barycenter of skeleton. Whereafter, elastic registration based on radial basis function are processed according to feature points of skeleton. Result of example demonstrate that according to traditional rigid registration, elastic registration based on skeleton feature retain natural difference in shape for organ's different part, and eliminate slight elastic deformation between frames caused by image obtained process simultaneously. This algorithm has a high practical value for image registration in course of 3D ultrasound image reconstruction.


    Guan Yepeng; Gu Weikang; Ye Xiuqing; Liu Jilin


    An algorithm for automatically extracting feature points is developed after the area of feature points in 2-dimensional (2D) imagebeing located by probability theory, correlated methods and criterion for abnormity. Feature points in 2D image can be extracted only by calculating standard deviation of gray within sampled pixels area in our approach statically. While extracting feature points, the limitation to confirm threshold by tentative method according to some a priori information on processing image can be avoided. It is proved that the proposed algorithm is valid and reliable by extracting feature points on actual natural images with abundant and weak texture, including multi-object with complex background, respectively. It can meet the demand of extracting feature points of 2D image automatically in machine vision system.


    K. Seetharaman


    Full Text Available This paper proposes a novel technique based on feature fusion using multiple linear regression analysis, and the least-square estimation method is employed to estimate the parameters. The given input query image is segmented into various regions according to the structure of the image. The color and texture features are extracted on each region of the query image, and the features are fused together using the multiple linear regression model. The estimated parameters of the model, which is modeled based on the features, are formed as a vector called a feature vector. The Canberra distance measure is adopted to compare the feature vectors of the query and target images. The F-measure is applied to evaluate the performance of the proposed technique. The obtained results expose that the proposed technique is comparable to the other existing techniques.

  11. Non-rigid registration of medical images based on ordinal feature and manifold learning

    Li, Qi; Liu, Jin; Zang, Bo


    With the rapid development of medical imaging technology, medical image research and application has become a research hotspot. This paper offers a solution to non-rigid registration of medical images based on ordinal feature (OF) and manifold learning. The structural features of medical images are extracted by combining ordinal features with local linear embedding (LLE) to improve the precision and speed of the registration algorithm. A physical model based on manifold learning and optimization search is constructed according to the complicated characteristics of non-rigid registration. The experimental results demonstrate the robustness and applicability of the proposed registration scheme.

  12. Image Analysis of Soil Micromorphology: Feature Extraction, Segmentation, and Quality Inference

    Petros Maragos


    Full Text Available We present an automated system that we have developed for estimation of the bioecological quality of soils using various image analysis methodologies. Its goal is to analyze soilsection images, extract features related to their micromorphology, and relate the visual features to various degrees of soil fertility inferred from biochemical characteristics of the soil. The image methodologies used range from low-level image processing tasks, such as nonlinear enhancement, multiscale analysis, geometric feature detection, and size distributions, to object-oriented analysis, such as segmentation, region texture, and shape analysis.

  13. Caroli's disease: magnetic resonance imaging features

    Guy, France; Cognet, Francois; Dranssart, Marie; Cercueil, Jean-Pierre; Conciatori, Laurent; Krause, Denis [Department of Radiology and Imaging, Dijon Le Bocage University Hospital, 2 Blvd. Marechal de Lattre de Tassigny, BP 1542, 21034 Dijon Cedex (France)


    Our objective was to describe the main aspects of MR imaging in Caroli's disease. Magnetic resonance cholangiography with a dynamic contrast-enhanced study was performed in nine patients with Caroli's disease. Bile duct abnormalities, lithiasis, dot signs, hepatic enhancement, renal abnormalities, and evidence of portal hypertension were evaluated. Three MR imaging patterns of Caroli's disease were found. In all but two patients, MR imaging findings were sufficient to confirm the diagnosis. Moreover, MR imaging provided information about the severity, location, and extent of liver involvement. This information was useful in planning the best therapeutic strategy. Magnetic resonance cholangiography with a dynamic contrast-enhanced study is a good screening tool for Caroli's disease. Direct cholangiography should be reserved for confirming doubtful cases. (orig.)

  14. clinical features and patterns of imaging in cerebral venous sinus ...


    Sep 1, 2013 ... study was conducted at Kenyatta National Hospital. Clinical and imaging records ... loss of memory, abdominal pain and senile dementia. Aetiological factors .... with a large study by Khealani in Pakistan and. Middle East that ...

  15. An Image Retrieval Method Based on Color and Texture Features


    The technique of image retrieval is widely used in science experiment, military affairs, public security,advertisement, family entertainment, library and so on. The existing algorithms are mostly based on the characteristics of color, texture, shape and space relationship. This paper introduced an image retrieval algorithm, which is based on the matching of weighted EMD(Earth Mover's Distance) distance and texture distance. EMD distance is the distance between the histograms of two images in HSV(Hue, Saturation, Value) color space, and texture distance is the L1 distance between the texture spectra of two images. The experimental results show that the retrieval rate can be increased obviously by using the proposed algorithm.

  16. Examplers based image fusion features for face recognition

    James, Alex Pappachen


    Examplers of a face are formed from multiple gallery images of a person and are used in the process of classification of a test image. We incorporate such examplers in forming a biologically inspired local binary decisions on similarity based face recognition method. As opposed to single model approaches such as face averages the exampler based approach results in higher recognition accu- racies and stability. Using multiple training samples per person, the method shows the following recognition accuracies: 99.0% on AR, 99.5% on FERET, 99.5% on ORL, 99.3% on EYALE, 100.0% on YALE and 100.0% on CALTECH face databases. In addition to face recognition, the method also detects the natural variability in the face images which can find application in automatic tagging of face images.

  17. Change detection in high resolution SAR images based on multiscale texture features

    Wen, Caihuan; Gao, Ziqiang


    This paper studied on change detection algorithm of high resolution (HR) Synthetic Aperture Radar (SAR) images based on multi-scale texture features. Firstly, preprocessed multi-temporal Terra-SAR images were decomposed by 2-D dual tree complex wavelet transform (DT-CWT), and multi-scale texture features were extracted from those images. Then, log-ratio operation was utilized to get difference images, and the Bayes minimum error theory was used to extract change information from difference images. Lastly, precision assessment was done. Meanwhile, we compared with the result of method based on texture features extracted from gray-level cooccurrence matrix (GLCM). We had a conclusion that, change detection algorithm based on multi-scale texture features has a great more improvement, which proves an effective method to change detect of high spatial resolution SAR images.

  18. Interpretation of archaeological small-scale features in spectral images

    Grøn, Ole; Palmer, Susanna; Stylegar, Frans-Arne;


    The paper's focus is the use of spectral images for the distinction of small archaeological anomalies on the basis of the authors work. Special attention is given to the ground-truthing perspective in the discussion of a number of cases from Norway. Different approaches to pattern-recognition are......-recognition are considered in the light of the increasing availability of hyper-spectral images that are difficult to analyse using visual inspection alone....

  19. New feature of the neutron color image intensifier

    Nittoh, Koichi; Konagai, Chikara; Noji, Takashi; Miyabe, Keisuke


    We developed prototype neutron color image intensifiers with high-sensitivity, wide dynamic range and long-life characteristics. In the prototype intensifier (Gd-Type 1), a terbium-activated Gd 2O 2S is used as the input-screen phosphor. In the upgraded model (Gd-Type 2), Gd 2O 3 and CsI:Na are vacuum deposited to form the phosphor layer, which improved the sensitivity and the spatial uniformity. A europium-activated Y 2O 2S multi-color scintillator, emitting red, green and blue photons with different intensities, is utilized as the output screen of the intensifier. By combining this image intensifier with a suitably tuned high-sensitive color CCD camera, higher sensitivity and wider dynamic range could be simultaneously attained than that of the conventional P20-phosphor-type image intensifier. The results of experiments at the JRR-3M neutron radiography irradiation port (flux: 1.5×10 8 n/cm 2/s) showed that these neutron color image intensifiers can clearly image dynamic phenomena with a 30 frame/s video picture. It is expected that the color image intensifier will be used as a new two-dimensional neutron sensor in new application fields.

  20. Large Margin Multi-Modal Multi-Task Feature Extraction for Image Classification.

    Yong Luo; Yonggang Wen; Dacheng Tao; Jie Gui; Chao Xu


    The features used in many image analysis-based applications are frequently of very high dimension. Feature extraction offers several advantages in high-dimensional cases, and many recent studies have used multi-task feature extraction approaches, which often outperform single-task feature extraction approaches. However, most of these methods are limited in that they only consider data represented by a single type of feature, even though features usually represent images from multiple modalities. We, therefore, propose a novel large margin multi-modal multi-task feature extraction (LM3FE) framework for handling multi-modal features for image classification. In particular, LM3FE simultaneously learns the feature extraction matrix for each modality and the modality combination coefficients. In this way, LM3FE not only handles correlated and noisy features, but also utilizes the complementarity of different modalities to further help reduce feature redundancy in each modality. The large margin principle employed also helps to extract strongly predictive features, so that they are more suitable for prediction (e.g., classification). An alternating algorithm is developed for problem optimization, and each subproblem can be efficiently solved. Experiments on two challenging real-world image data sets demonstrate the effectiveness and superiority of the proposed method.

  1. Nonlinear Second-Order Partial Differential Equation-Based Image Smoothing Technique

    Tudor Barbu


    Full Text Available A second-order nonlinear parabolic PDE-based restoration model is provided in this article. The proposed anisotropic diffusion-based denoising approach is based on some robust versions of the edge-stopping function and of the conductance parameter. Two stable and consistent approximation schemes are then developed for this differential model. Our PDE-based filtering technique achieves an efficient noise removal while preserving the edges and other image features. It outperforms both the conventional filters and also many PDE-based denoising approaches, as it results from the successful experiments and method comparison applied.

  2. Automatic Tissue Differentiation Based on Confocal Endomicroscopic Images for Intraoperative Guidance in Neurosurgery

    Ali Kamen


    Full Text Available Diagnosis of tumor and definition of tumor borders intraoperatively using fast histopathology is often not sufficiently informative primarily due to tissue architecture alteration during sample preparation step. Confocal laser microscopy (CLE provides microscopic information of tissue in real-time on cellular and subcellular levels, where tissue characterization is possible. One major challenge is to categorize these images reliably during the surgery as quickly as possible. To address this, we propose an automated tissue differentiation algorithm based on the machine learning concept. During a training phase, a large number of image frames with known tissue types are analyzed and the most discriminant image-based signatures for various tissue types are identified. During the procedure, the algorithm uses the learnt image features to assign a proper tissue type to the acquired image frame. We have verified this method on the example of two types of brain tumors: glioblastoma and meningioma. The algorithm was trained using 117 image sequences containing over 27 thousand images captured from more than 20 patients. We achieved an average cross validation accuracy of better than 83%. We believe this algorithm could be a useful component to an intraoperative pathology system for guiding the resection procedure based on cellular level information.

  3. Spectrum Feature Retrieval and Comparison of Remote Sensing Images Using Improved ISODATA Algorithm

    刘磊; 敬忠良; 肖刚


    Due to the large quantities of data and high relativity of the spectra of remote sensing images, K-L transformation is used to eliminate the relativity. An improved ISODATA(Interative Self-Organizing Data Analysis Technique A) algorithm is used to extract the spectrum features of the images. The computation is greatly reduced and the dynamic arguments are realized. The comparison of features between two images is carried out, and good results are achieved in simulation.

  4. Differential Spatio-temporal Multiband Satellite Image Clustering using K-means Optimization With Reinforcement Programming

    Irene Erlyn Wina Rachmawan


    Full Text Available Deforestration is one of the crucial issues in Indonesia because now Indonesia has world's highest deforestation rate. In other hand, multispectral image delivers a great source of data for studying spatial and temporal changeability of the environmental such as deforestration area. This research present differential image processing methods for detecting nature change of deforestration. Our differential image processing algorithms extract and indicating area automatically. The feature of our proposed idea produce extracted information from multiband satellite image and calculate the area of deforestration by years with calculating data using temporal dataset. Yet, multiband satellite image consists of big data size that were difficult to be handled for segmentation. Commonly, K- Means clustering is considered to be a powerfull clustering algorithm because of its ability to clustering big data. However K-Means has sensitivity of its first generated centroids, which could lead into a bad performance. In this paper we propose a new approach to optimize K-Means clustering using Reinforcement Programming in order to clustering multispectral image. We build a new mechanism for generating initial centroids by implementing exploration and exploitation knowledge from Reinforcement Programming. This optimization will lead a better result for K-means data cluster. We select multispectral image from Landsat 7 in past ten years in Medawai, Borneo, Indonesia, and apply two segmentation areas consist of deforestration land and forest field. We made series of experiments and compared the experimental results of K-means using Reinforcement Programming as optimizing initiate centroid and normal K-means without optimization process. Keywords: Deforestration, Multispectral images, landsat, automatic clustering, K-means.

  5. Hyperspectral Image Classification Based on the Weighted Probabilistic Fusion of Multiple Spectral-spatial Features

    ZHANG Chunsen


    Full Text Available A hyperspectral images classification method based on the weighted probabilistic fusion of multiple spectral-spatial features was proposed in this paper. First, the minimum noise fraction (MNF approach was employed to reduce the dimension of hyperspectral image and extract the spectral feature from the image, then combined the spectral feature with the texture feature extracted based on gray level co-occurrence matrix (GLCM, the multi-scale morphological feature extracted based on OFC operator and the end member feature extracted based on sequential maximum angle convex cone (SMACC method to form three spectral-spatial features. Afterwards, support vector machine (SVM classifier was used for the classification of each spectral-spatial feature separately. Finally, we established the weighted probabilistic fusion model and applied the model to fuse the SVM outputs for the final classification result. In order to verify the proposed method, the ROSIS and AVIRIS image were used in our experiment and the overall accuracy reached 97.65% and 96.62% separately. The results indicate that the proposed method can not only overcome the limitations of traditional single-feature based hyperspectral image classification, but also be superior to conventional VS-SVM method and probabilistic fusion method. The classification accuracy of hyperspectral images was improved effectively.

  6. Differentiation of intrahepatic mass-forming cholangiocarcinoma from hepatocellular carcinoma on gadoxetic acid-enhanced liver MR imaging

    Kim, Rihyeon; Shin, Cheong-Il; Yoon, Jeong Hee; Joo, Ijin; Kim, Seong Ho; Hwang, Inpyeong [Seoul National University Hospital, Department of Radiology, Seoul (Korea, Republic of); Lee, Jeong Min; Han, Joon Koo [Seoul National University Hospital, Department of Radiology, Seoul (Korea, Republic of); Seoul National University Hospital, Institute of Radiation Medicine, Seoul (Korea, Republic of); Lee, Eun Sun; Choi, Byung Ihn [Chung-Ang University Hospital, Department of Radiology, Seoul (Korea, Republic of)


    To determine the different imaging features of intrahepatic mass-forming cholangiocarcinoma (IMCC) from hepatocellular carcinoma (HCC) on gadoxetic acid-enhanced magnetic resonance imaging (MRI). This retrospective study was institutional review board approved and the requirement for informed consent was waived. Patients who underwent gadoxetic acid-enhanced MRI with histologically confirmed IMCCs (n = 46) or HCCs (n = 58) were included. Imaging features of IMCCs and HCCs on gadoxetic acid-enhanced MRI including T2- and T1-weighted, diffusion weighted images, dynamic study and hepatobiliary phase (HBP) images were analyzed. Univariate and multivariate logistic regression analyses were performed to identify relevant differentiating features between IMCCs and HCCs. Multivariate analysis revealed heterogeneous T2 signal intensity and a hypointense rim on the HBP as suggestive findings of IMCCs and the wash-in and ''portal wash-out'' enhancement pattern as well as focal T1 high signal intensity foci as indicative of HCCs (all, p < 0.05). When we combined any three of the above four imaging features, we were able to diagnose IMCCs with 94 % (43/46) sensitivity and 86 % (50/58) specificity. Combined interpretation of enhancement characteristics including HBP images, morphologic features, and strict application of the ''portal wash-out'' pattern helped more accurate discrimination of IMCCs from HCCs. (orig.)

  7. Skin cancer texture analysis of OCT images based on Haralick, fractal dimension and the complex directional field features

    Raupov, Dmitry S.; Myakinin, Oleg O.; Bratchenko, Ivan A.; Kornilin, Dmitry V.; Zakharov, Valery P.; Khramov, Alexander G.


    Optical coherence tomography (OCT) is usually employed for the measurement of tumor topology, which reflects structural changes of a tissue. We investigated the possibility of OCT in detecting changes using a computer texture analysis method based on Haralick texture features, fractal dimension and the complex directional field method from different tissues. These features were used to identify special spatial characteristics, which differ healthy tissue from various skin cancers in cross-section OCT images (B-scans). Speckle reduction is an important pre-processing stage for OCT image processing. In this paper, an interval type-II fuzzy anisotropic diffusion algorithm for speckle noise reduction in OCT images was used. The Haralick texture feature set includes contrast, correlation, energy, and homogeneity evaluated in different directions. A box-counting method is applied to compute fractal dimension of investigated tissues. Additionally, we used the complex directional field calculated by the local gradient methodology to increase of the assessment quality of the diagnosis method. The complex directional field (as well as the "classical" directional field) can help describe an image as set of directions. Considering to a fact that malignant tissue grows anisotropically, some principal grooves may be observed on dermoscopic images, which mean possible existence of principal directions on OCT images. Our results suggest that described texture features may provide useful information to differentiate pathological from healthy patients. The problem of recognition melanoma from nevi is decided in this work due to the big quantity of experimental data (143 OCT-images include tumors as Basal Cell Carcinoma (BCC), Malignant Melanoma (MM) and Nevi). We have sensitivity about 90% and specificity about 85%. Further research is warranted to determine how this approach may be used to select the regions of interest automatically.

  8. Feature Understanding and Target Detection for Sparse Microwave Synthetic Aperture Radar Images

    Zhang Zenghui


    Full Text Available Sparse microwave imaging using sparse priors of observed scenes in space, time, frequency, or polarization domain and echo data with sampling rate smaller than the traditional Nyquist rate as well as optimization algorithms for reconstructing the microwave images of observed scenes has many advantages over traditional microwave imaging systems. In sparse microwave imaging, image acquisition and representation vary; therefore, new feature analysis and cognitive interpretation theories and methods should be developed based on current research results. In this study, we analyze the statistical properties of sparse Synthetic Aperture Radar (SAR images and changes in point, line and regional features induced by sparse reconstruction. For SAR images recovered by the spatial sparse model, the statistical distribution degrades, whereas points and lines can be accurately extracted by low sampling rates. Furthermore, the target detection method based on sparse SAR images is studied. Owing to a weak background noise, target detection is easier using sparse SAR images than traditional ones.

  9. Automatic Image Registration of Multi-Modal Remotely Sensed Data with Global Shearlet Features

    Murphy, James M.; Le Moigne, Jacqueline; Harding, David J.


    Automatic image registration is the process of aligning two or more images of approximately the same scene with minimal human assistance. Wavelet-based automatic registration methods are standard, but sometimes are not robust to the choice of initial conditions. That is, if the images to be registered are too far apart relative to the initial guess of the algorithm, the registration algorithm does not converge or has poor accuracy, and is thus not robust. These problems occur because wavelet techniques primarily identify isotropic textural features and are less effective at identifying linear and curvilinear edge features. We integrate the recently developed mathematical construction of shearlets, which is more effective at identifying sparse anisotropic edges, with an existing automatic wavelet-based registration algorithm. Our shearlet features algorithm produces more distinct features than wavelet features algorithms; the separation of edges from textures is even stronger than with wavelets. Our algorithm computes shearlet and wavelet features for the images to be registered, then performs least squares minimization on these features to compute a registration transformation. Our algorithm is two-staged and multiresolution in nature. First, a cascade of shearlet features is used to provide a robust, though approximate, registration. This is then refined by registering with a cascade of wavelet features. Experiments across a variety of image classes show an improved robustness to initial conditions, when compared to wavelet features alone.

  10. Detection and Classification of Cancer from Microscopic Biopsy Images Using Clinically Significant and Biologically Interpretable Features

    Kumar, Rajesh; Srivastava, Subodh


    A framework for automated detection and classification of cancer from microscopic biopsy images using clinically significant and biologically interpretable features is proposed and examined. The various stages involved in the proposed methodology include enhancement of microscopic images, segmentation of background cells, features extraction, and finally the classification. An appropriate and efficient method is employed in each of the design steps of the proposed framework after making a comparative analysis of commonly used method in each category. For highlighting the details of the tissue and structures, the contrast limited adaptive histogram equalization approach is used. For the segmentation of background cells, k-means segmentation algorithm is used because it performs better in comparison to other commonly used segmentation methods. In feature extraction phase, it is proposed to extract various biologically interpretable and clinically significant shapes as well as morphology based features from the segmented images. These include gray level texture features, color based features, color gray level texture features, Law's Texture Energy based features, Tamura's features, and wavelet features. Finally, the K-nearest neighborhood method is used for classification of images into normal and cancerous categories because it is performing better in comparison to other commonly used methods for this application. The performance of the proposed framework is evaluated using well-known parameters for four fundamental tissues (connective, epithelial, muscular, and nervous) of randomly selected 1000 microscopic biopsy images. PMID:27006938

  11. Use of Wavelet-Fuzzy Features with PCA for Image Registration

    Safia Sadruddin


    Full Text Available In this paper, we discuss an Image Registration system based on neural network, which uses Wavelet-fuzzy features of an image. In this system, Wavelet-fuzzy features are extracted from an image and then reduced using Principal Component Analysis (PCA. The reduced feature set is then used for training the neural network for image registration. The geometric transformation between the reference and sensed image sets are evaluated using affine transformation parameters. The trained neural network produces registration parameters (translation, rotation and scaling with respect to reference and sensed image. Two parameters namely Mean Absolute Registration Error and Mutual Information are used as evaluation parameters. Experimentally, we show that the proposed technique for image registration is accurate and robust for distorted and noisy inputs.

  12. The Machine Recognition for Population Feature of Wheat Images Based on BP Neural Network

    LI Shao-kun; SUO Xing-mei; BAI Zhong-ying; QI Zhi-li; Liu Xiao-hong; GAO Shi-ju; ZHAO Shuang-ning


    Recognition and analysis of dynamic information about population images during wheat growth periods can be taken for the base of quantitative diagnosis for wheat growth. A recognition system based on self-learning BP neural network for feature data of wheat population images, such as total green areas and leaves areas was designed in this paper. In addition, some techniques to create favorable conditions for image recognition was discussed, which were as follows: (1) The method of collecting images by a digital camera and assistant equipment under natural conditions in fields. (2) An algorithm of pixei labeling was used to segment image and extract feature. (3)A high pass filter based on Laplacian was used to strengthen image information. The results showed that the ANN system was availability for image recognition of wheat population feature.

  13. Improved Feature Detection in Fused Intensity-Range Images with Complex SIFT (ℂSIFT

    Boris Jutzi


    Full Text Available The real and imaginary parts are proposed as an alternative to the usual Polar representation of complex-valued images. It is proven that the transformation from Polar to Cartesian representation contributes to decreased mutual information, and hence to greater distinctiveness. The Complex Scale-Invariant Feature Transform (ℂSIFT detects distinctive features in complex-valued images. An evaluation method for estimating the uniformity of feature distributions in complex-valued images derived from intensity-range images is proposed. In order to experimentally evaluate the proposed methodology on intensity-range images, three different kinds of active sensing systems were used: Range Imaging, Laser Scanning, and Structured Light Projection devices (PMD CamCube 2.0, Z+F IMAGER 5003, Microsoft Kinect.

  14. Featured Image: A Galaxy Plunges Into a Cluster Core

    Kohler, Susanna


    The galaxy that takes up most of the frame in this stunning image (click for the full view!) is NGC 1427A. This is a dwarf irregular galaxy (unlike the fortuitously-located background spiral galaxy in the lower right corner of the image), and its currently in the process of plunging into the center of the Fornax galaxy cluster. Marcelo Mora (Pontifical Catholic University of Chile) and collaborators have analyzed observations of this galaxy made by both the Very Large Telescope in Chile and the Hubble Advanced Camera for Surveys, which produced the image shown here as a color composite in three channels. The team worked to characterize the clusters of star formation within NGC 1427A identifiable in the image as bright knots within the galaxy and determine how the interactions of this galaxy with its cluster environment affect the star formation within it. For more information and the original image, see the paper below.Citation:Marcelo D. Mora et al 2015 AJ 150 93. doi:10.1088/0004-6256/150/3/93

  15. MR imaging features of foot involvement in patients with psoriasis

    Erdem, C. Zuhal [Department of Radiology, Zonguldak Karaelmas University, School of Medicine, Zonguldak (Turkey)], E-mail:; Tekin, Nilgun Solak [Department of Dermatology, Zonguldak Karaelmas University, School of Medicine, Zonguldak (Turkey); Sarikaya, Selda [Department of Physical Therapy and Rehabilitation, Zonguldak Karaelmas University, School of Medicine, Zonguldak (Turkey); Erdem, L. Oktay; Gulec, Sezen [Department of Radiology, Zonguldak Karaelmas University, School of Medicine, Zonguldak (Turkey)


    Objective: To determine alterations of the soft tissues, tendons, cartilage, joint spaces, and bones of the foot using magnetic resonance (MR) imaging in patients with psoriasis. Materials and methods: Clinical and MR examination of the foot was performed in 26 consecutive patients (52 ft) with psoriasis. As a control group, 10 healthy volunteers (20 ft) were also studied. Joint effusion/synovitis, retrocalcaneal bursitis, retroachilles bursitis, Achilles tendonitis, soft-tissue edema, para-articular enthesophytes, bone marrow edema, sinus tarsi syndrome, enthesopathy at the Achilles attachment and at the plantar fascia attachment, plantar fasciitis, tenosynovitis, subchondral cysts, and bone erosions, joint space narrowing, subchondral signal changes, osteolysis, luxation, and sub-luxation were examined. Results: Clinical signs and symptoms (pain and swelling) due to foot involvement were present in none of the patients while frequency of involvement was 92% (24/26) by MR imaging. The most common MR imaging findings were Achilles tendonitis (acute and peritendinitis) (57%), retrocalcaneal bursitis (50%), joint effusion/synovitis (46%), soft-tissue edema (46%), and para-articular enthesophytes (38%). The most commonly involved anatomical region was the hindfoot (73%). Conclusion: Our data showed that the incidence of foot involvement was very high in asymptomatic patients with psoriasis on MR imaging. Further MR studies are needed to confirm these data. We conclude that MR imaging may be of importance especially in early diagnosis and treatment of inflammatory changes in the foot.

  16. Color and Texture Feature for Remote Sensing - Image Retrieval System: A Comparative Study

    Retno Kusumaningrum


    Full Text Available In this study, we proposed score fusion technique to improve the performance of remote sensing image retrieval system (RS-IRS using combination of several features. The representation of each feature is selected based on their performance when used as single feature in RS-IRS. Those features are color moment using L*a*b* color space, edge direction histogram extracted from Saturation channel, GLCM and Gabor Wavelet represented using standard deviation, and local binary pattern using 8-neighborhood. The score fusion is performed by computing the value of image similarity between an image in the database and query, where the image similarity value is sum of all features similarity, where each of feature similarity has been divided by SVD value of feature similarity between all images in the database and query from related feature. The feature similarity is measured by histogram intersection for local binary pattern, whereas the color moment, edge direction histogram, GLCM, and Gabor are measured by Euclidean Distance. The final result shows that the best performance of remote sensing image retrieval in this study is a system which uses the combination of color and texture features (i.e. color moment, edge direction histogram, GLCM, Gabor wavelet, and local binary pattern and uses score fusion in measuring the image similarity between query and images in the database. This system outperforms the other five individual feature with average precision rates 3%, 20%, 13%, 11%, and 9%, respectively, for color moment, edge direction histogram, GLCM, Gabor wavelet, and LBP. Moreover, this system also increase 17% compared to system without score fusion, simple-sum technique.

  17. Ensemble classification of colon biopsy images based on information rich hybrid features.

    Rathore, Saima; Hussain, Mutawarra; Aksam Iftikhar, Muhammad; Jalil, Abdul


    In recent years, classification of colon biopsy images has become an active research area. Traditionally, colon cancer is diagnosed using microscopic analysis. However, the process is subjective and leads to considerable inter/intra observer variation. Therefore, reliable computer-aided colon cancer detection techniques are in high demand. In this paper, we propose a colon biopsy image classification system, called CBIC, which benefits from discriminatory capabilities of information rich hybrid feature spaces, and performance enhancement based on ensemble classification methodology. Normal and malignant colon biopsy images differ with each other in terms of the color distribution of different biological constituents. The colors of different constituents are sharp in normal images, whereas the colors diffuse with each other in malignant images. In order to exploit this variation, two feature types, namely color components based statistical moments (CCSM) and Haralick features have been proposed, which are color components based variants of their traditional counterparts. Moreover, in normal colon biopsy images, epithelial cells possess sharp and well-defined edges. Histogram of oriented gradients (HOG) based features have been employed to exploit this information. Different combinations of hybrid features have been constructed from HOG, CCSM, and Haralick features. The minimum Redundancy Maximum Relevance (mRMR) feature selection method has been employed to select meaningful features from individual and hybrid feature sets. Finally, an ensemble classifier based on majority voting has been proposed, which classifies colon biopsy images using the selected features. Linear, RBF, and sigmoid SVM have been employed as base classifiers. The proposed system has been tested on 174 colon biopsy images, and improved performance (=98.85%) has been observed compared to previously reported studies. Additionally, the use of mRMR method has been justified by comparing the

  18. Similar Reference Image Quality Assessment: A New Database and A Trial with Local Feature Matching

    Lu, Qingbo; Zhou, Wengang; Li, Houqiang


    Conventionally, the reference image for image quality assessment (IQA) is completely available (full-reference IQA) or unavailable (no-reference IQA). Even for reduced-reference IQA, the features that are used to predict image quality are still extracted from the pristine reference image. However, the pristine reference image is always unavailable in many real scenarios. In contrast, it is convenient to obtain a number of similar reference images via retrieval from the Internet. These similar reference images may share similar contents and scenes with the image to be assessed. In this paper, we attempt to discuss the image quality assessment problem from the view of similar images, i.e. similar reference IQA. Although the similar reference images share similar contents with the degraded image, the difference between them still cannot be ignored. Therefore, we propose an IQA framework based on local feature matching, which can help to identify the similar regions and structures. Then the IQA features are computed only from these similar regions to predict the final image quality score. Besides, since there is no IQA databases for the similar reference IQA problem, we establish a novel IQA database that consists of 272 images from four scenes. The experiments demonstrate that the performance of our scheme goes beyond state-of-the-art no-reference IQA methods and some full-reference IQA algorithms.

  19. Image indexing using composite color and shape invariant features

    Gevers, Th.; Smeulders, A.W.M.


    New sets of color models are proposed for object recognition invariant to a change in view point, object geometry and illumination. Further, computational methods are presented to combine color and shape invariants to produce a high-dimensional invariant feature set for discriminatory object recogni

  20. A new method to extract stable feature points based on self-generated simulation images

    Long, Fei; Zhou, Bin; Ming, Delie; Tian, Jinwen


    Recently, image processing has got a lot of attention in the field of photogrammetry, medical image processing, etc. Matching two or more images of the same scene taken at different times, by different cameras, or from different viewpoints, is a popular and important problem. Feature extraction plays an important part in image matching. Traditional SIFT detectors reject the unstable points by eliminating the low contrast and edge response points. The disadvantage is the need to set the threshold manually. The main idea of this paper is to get the stable extremums by machine learning algorithm. Firstly we use ASIFT approach coupled with the light changes and blur to generate multi-view simulated images, which make up the set of the simulated images of the original image. According to the way of generating simulated images set, affine transformation of each generated image is also known. Instead of the traditional matching process which contain the unstable RANSAC method to get the affine transformation, this approach is more stable and accurate. Secondly we calculate the stability value of the feature points by the set of image with its affine transformation. Then we get the different feature properties of the feature point, such as DOG features, scales, edge point density, etc. Those two form the training set while stability value is the dependent variable and feature property is the independent variable. At last, a process of training by Rank-SVM is taken. We will get a weight vector. In use, based on the feature properties of each points and weight vector calculated by training, we get the sort value of each feature point which refers to the stability value, then we sort the feature points. In conclusion, we applied our algorithm and the original SIFT detectors to test as a comparison. While in different view changes, blurs, illuminations, it comes as no surprise that experimental results show that our algorithm is more efficient.

  1. The relationship study between image features and detection probability based on psychology experiments

    Lin, Wei; Chen, Yu-hua; Wang, Ji-yuan; Gao, Hong-sheng; Wang, Ji-jun; Su, Rong-hua; Mao, Wei


    Detection probability is an important index to represent and estimate target viability, which provides basis for target recognition and decision-making. But it will expend a mass of time and manpower to obtain detection probability in reality. At the same time, due to the different interpretation of personnel practice knowledge and experience, a great difference will often exist in the datum obtained. By means of studying the relationship between image features and perception quantity based on psychology experiments, the probability model has been established, in which the process is as following.Firstly, four image features have been extracted and quantified, which affect directly detection. Four feature similarity degrees between target and background were defined. Secondly, the relationship between single image feature similarity degree and perception quantity was set up based on psychological principle, and psychological experiments of target interpretation were designed which includes about five hundred people for interpretation and two hundred images. In order to reduce image features correlativity, a lot of artificial synthesis images have been made which include images with single brightness feature difference, images with single chromaticity feature difference, images with single texture feature difference and images with single shape feature difference. By analyzing and fitting a mass of experiments datum, the model quantitys have been determined. Finally, by applying statistical decision theory and experimental results, the relationship between perception quantity with target detection probability has been found. With the verification of a great deal of target interpretation in practice, the target detection probability can be obtained by the model quickly and objectively.

  2. Relationship between Hyperuricemia and Haar-Like Features on Tongue Images

    Yan Cui


    Full Text Available Objective. To investigate differences in tongue images of subjects with and without hyperuricemia. Materials and Methods. This population-based case-control study was performed in 2012-2013. We collected data from 46 case subjects with hyperuricemia and 46 control subjects, including results of biochemical examinations and tongue images. Symmetrical Haar-like features based on integral images were extracted from tongue images. T-tests were performed to determine the ability of extracted features to distinguish between the case and control groups. We first selected features using the common criterion P<0.05, then conducted further examination of feature characteristics and feature selection using means and standard deviations of distributions in the case and control groups. Results. A total of 115,683 features were selected using the criterion P<0.05. The maximum area under the receiver operating characteristic curve (AUC of these features was 0.877. The sensitivity of the feature with the maximum AUC value was 0.800 and specificity was 0.826 when the Youden index was maximized. Features that performed well were concentrated in the tongue root region. Conclusions. Symmetrical Haar-like features enabled discrimination of subjects with and without hyperuricemia in our sample. The locations of these discriminative features were in agreement with the interpretation of tongue appearance in traditional Chinese and Western medicine.

  3. Malformations of cortical development: 3T magnetic resonance imaging features

    Battal, Bilal; Ince, Selami; Akgun, Veysel; Kocaoglu, Murat; Ozcan, Emrah; Tasar, Mustafa


    Malformation of cortical development (MCD) is a term representing an inhomogeneous group of central nervous system abnormalities, referring particularly to embriyological aspect as a consequence of any of the three developmental stages, i.e., cell proliferation, cell migration and cortical organization. These include cotical dysgenesis, microcephaly, polymicrogyria, schizencephaly, lissencephaly, hemimegalencephaly, heterotopia and focal cortical dysplasia. Since magnetic resonance imaging is the modality of choice that best identifies the structural anomalies of the brain cortex, we aimed to provide a mini review of MCD by using 3T magnetic resonance scanner images. PMID:26516429

  4. Feature exploration for biometric recognition using millimetre wave body images


    The electronic version of this article is the complete one and can be found online at: The use of millimetre wave images has been proposed recently in the biometric field to overcome certain limitations when using images acquired at visible frequencies. Furthermore, the security community has started using millimetre wave screening scanners in order to detect concealed objects. We believe we can exploit the use of these devices by incorporating b...

  5. Malformations of cortical development:3T magnetic resonance imaging features

    Bilal; Battal; Selami; Ince; Veysel; Akgun; Murat; Kocaoglu; Emrah; Ozcan; Mustafa; Tasar


    Malformation of cortical development(MCD) is a term representing an inhomogeneous group of central nervous system abnormalities, referring particularly to embriyological aspect as a consequence of any of the three developmental stages, i.e., cell proliferation, cell migration and cortical organization. These include cotical dysgenesis, microcephaly, polymicrogyria, schizencephaly, lissencephaly, hemimegalencephaly, heterotopia and focal cortical dysplasia. Since magnetic resonance imaging is the modality of choice that best identifies the structural anomalies of the brain cortex, we aimed to provide a mini review of MCD by using 3T magnetic resonance scanner images.

  6. X-ray image enhancement via determinant based feature selection.

    Tappenden, R; Hegarty, J; Broughton, R; Butler, A; Coope, I; Renaud, P


    Previous work has investigated the feasibility of using Eigenimage-based enhancement tools to highlight abnormalities on chest X-rays (Butler et al in J Med Imaging Radiat Oncol 52:244-253, 2008). While promising, this approach has been limited by computational restrictions of standard clinical workstations, and uncertainty regarding what constitutes an adequate sample size. This paper suggests an alternative mathematical model to the above referenced singular value decomposition method, which can significantly reduce both the required sample size and the time needed to perform analysis. Using this approach images can be efficiently separated into normal and abnormal parts, with the potential for rapid highlighting of pathology.

  7. Manifold-based feature point matching for multi-modal image registration.

    Hu, Liang; Wang, Manning; Song, Zhijian


    Images captured using different modalities usually have significant variations in their intensities, which makes it difficult to reveal their internal structural similarities and achieve accurate registration. Most conventional feature-based image registration techniques are fast and efficient, but they cannot be used directly for the registration of multi-modal images because of these intensity variations. This paper introduces the theory of manifold learning to transform the original images into mono-modal modalities, which is a feature-based method that is applicable to multi-modal image registration. Subsequently, scale-invariant feature transform is used to detect highly distinctive local descriptors and matches between corresponding images, and a point-based registration is executed. The algorithm was tested with T1- and T2-weighted magnetic resonance (MR) images obtained from BrainWeb. Both qualitative and quantitative evaluations of the method were performed and the results compared with those produced previously. The experiments showed that feature point matching after manifold learning achieved more accurate results than did the similarity measure for multi-modal image registration. This study provides a new manifold-based feature point matching method for multi-modal medical image registration, especially for MR images. The proposed method performs better than do conventional intensity-based techniques in terms of its registration accuracy and is suitable for clinical procedures. Copyright © 2012 John Wiley & Sons, Ltd.

  8. Medical image retrieval based on texture and shape feature co-occurrence

    Zhou, Yixiao; Huang, Yan; Ling, Haibin; Peng, Jingliang


    With the rapid development and wide application of medical imaging technology, explosive volumes of medical image data are produced every day all over the world. As such, it becomes increasingly challenging to manage and utilize such data effectively and efficiently. In particular, content-based medical image retrieval has been intensively researched in the past decade or so. In this work, we propose a novel approach to content-based medical image retrieval utilizing the co-occurrence of both the texture and the shape features in contrast to most previous algorithms that use purely the texture or the shape feature. Specifically, we propose a novel form of representation for the co-occurrence of the texture and the shape features in an image, i.e., the gray level and edge direction co-occurrence matrix (GLEDCOM). Based on GLEDCOM, we define eleven features forming a feature vector that is used to measure the similarity between images. As a result, it consistently yields outstanding performance on both images rich in texture (e.g., image of brain) and images with dominant smooth regions and sharp edges (e.g., image of bladder). As demonstrated by experiments, the mean precision of retrieval with GLEDCOM algorithm outperforms a set of representative algorithms including the gray level co-occurrence matrix (GLCM) based, the Hu's seven moment invariants (HSMI) based, the uniformity estimation method (UEM) based and the the modified Zernike moments (MZM) based algorithms by 10%-20%.

  9. Image processing based automatic diagnosis of glaucoma using wavelet features of segmented optic disc from fundus image.

    Singh, Anushikha; Dutta, Malay Kishore; ParthaSarathi, M; Uher, Vaclav; Burget, Radim


    Glaucoma is a disease of the retina which is one of the most common causes of permanent blindness worldwide. This paper presents an automatic image processing based method for glaucoma diagnosis from the digital fundus image. In this paper wavelet feature extraction has been followed by optimized genetic feature selection combined with several learning algorithms and various parameter settings. Unlike the existing research works where the features are considered from the complete fundus or a sub image of the fundus, this work is based on feature extraction from the segmented and blood vessel removed optic disc to improve the accuracy of identification. The experimental results presented in this paper indicate that the wavelet features of the segmented optic disc image are clinically more significant in comparison to features of the whole or sub fundus image in the detection of glaucoma from fundus image. Accuracy of glaucoma identification achieved in this work is 94.7% and a comparison with existing methods of glaucoma detection from fundus image indicates that the proposed approach has improved accuracy of classification.

  10. MicroRNAs are required for the feature maintenance and differentiation of brown adipocytes.

    Kim, Hye-Jin; Cho, Hyunjii; Alexander, Ryan; Patterson, Heide Christine; Gu, Minxia; Lo, Kinyui Alice; Xu, Dan; Goh, Vera J; Nguyen, Long N; Chai, Xiaoran; Huang, Cher X; Kovalik, Jean-Paul; Ghosh, Sujoy; Trajkovski, Mirko; Silver, David L; Lodish, Harvey; Sun, Lei


    Brown adipose tissue (BAT) is specialized to burn lipids for heat generation as a natural defense against cold and obesity. Previous studies established microRNAs (miRNAs) as essential regulators of brown adipocyte differentiation, but whether miRNAs are required for the feature maintenance of mature brown adipocytes remains unknown. To address this question, we ablated Dgcr8, a key regulator of the miRNA biogenesis pathway, in mature brown as well as in white adipocytes. Adipose tissue-specific Dgcr8 knockout mice displayed enlarged but pale interscapular brown fat with decreased expression of genes characteristic of brown fat and were intolerant to cold exposure. Primary brown adipocyte cultures in vitro confirmed that miRNAs are required for marker gene expression in mature brown adipocytes. We also demonstrated that miRNAs are essential for the browning of subcutaneous white adipocytes in vitro and in vivo. Using this animal model, we performed miRNA expression profiling analysis and identified a set of BAT-specific miRNAs that are upregulated during brown adipocyte differentiation and enriched in brown fat compared with other organs. We identified miR-182 and miR-203 as new regulators of brown adipocyte development. Taken together, our study demonstrates an essential role of miRNAs in the maintenance as well as in the differentiation of brown adipocytes.

  11. MicroRNAs Are Required for the Feature Maintenance and Differentiation of Brown Adipocytes

    Kim, Hye-Jin; Cho, Hyunjii; Alexander, Ryan; Patterson, Heide Christine; Gu, Minxia; Lo, Kinyui Alice; Xu, Dan; Goh, Vera J.; Nguyen, Long N.; Chai, Xiaoran; Huang, Cher X.; Kovalik, Jean-Paul; Ghosh, Sujoy; Trajkovski, Mirko; Silver, David L.; Lodish, Harvey


    Brown adipose tissue (BAT) is specialized to burn lipids for heat generation as a natural defense against cold and obesity. Previous studies established microRNAs (miRNAs) as essential regulators of brown adipocyte differentiation, but whether miRNAs are required for the feature maintenance of mature brown adipocytes remains unknown. To address this question, we ablated Dgcr8, a key regulator of the miRNA biogenesis pathway, in mature brown as well as in white adipocytes. Adipose tissue–specific Dgcr8 knockout mice displayed enlarged but pale interscapular brown fat with decreased expression of genes characteristic of brown fat and were intolerant to cold exposure. Primary brown adipocyte cultures in vitro confirmed that miRNAs are required for marker gene expression in mature brown adipocytes. We also demonstrated that miRNAs are essential for the browning of subcutaneous white adipocytes in vitro and in vivo. Using this animal model, we performed miRNA expression profiling analysis and identified a set of BAT-specific miRNAs that are upregulated during brown adipocyte differentiation and enriched in brown fat compared with other organs. We identified miR-182 and miR-203 as new regulators of brown adipocyte development. Taken together, our study demonstrates an essential role of miRNAs in the maintenance as well as in the differentiation of brown adipocytes. PMID:25008181

  12. Differential surface models for tactile perception of shape and on-line tracking of features

    Hemami, H.


    Tactile perception of shape involves an on-line controller and a shape perceptor. The purpose of the on-line controller is to maintain gliding or rolling contact with the surface, and collect information, or track specific features of the surface such as edges of a certain sharpness. The shape perceptor uses the information to perceive, estimate the parameters of, or recognize the shape. The differential surface model depends on the information collected and on the a priori information known about the robot and its physical parameters. These differential models are certain functionals that are projections of the dynamics of the robot onto the surface gradient or onto the tangent plane. A number of differential properties may be directly measured from present day tactile sensors. Others may have to be indirectly computed from measurements. Others may constitute design objectives for distributed tactile sensors of the future. A parameterization of the surface leads to linear and nonlinear sequential parameter estimation techniques for identification of the surface. Many interesting compromises between measurement and computation are possible.

  13. Global image feature extraction using slope pattern spectra

    Toudjeu, IT


    Full Text Available of coffee beans. Granulometries were also used to estimate the dominant width of the white patterns in the X-ray images of welds [7]. Due to the computational load associated with the calculation of granulometries, Vincent [6], building on the work...

  14. Magnetic resonance imaging features of chloroma of the shoulder

    Gomez, N. [Department of Radiology, Hospital de la Princesa, Madrid (Spain)]|[Department of Radiology, Hospital de la Princesa, Madrid (Spain); Ocon, E. [Department of Radiology, Hospital de la Princesa, Madrid (Spain); Friera, A. [Department of Radiology, Hospital de la Princesa, Madrid (Spain); Penarrubia, M.J. [Department of Hematology, Hospital de la Princesa, Madrid (Spain); Acevedo, A. [Department of Pathology, Hospital de la Princesa, Madrid (Spain)


    A patient with a history of essential thrombocytosis presented with diffuse skeletal pain and restricted motion of the left shoulder. Magnetic resonance imaging (MRI) of the left glenohumeral joint showed a soft tissue mass that displaced the rotator cuff. Biopsy of the mass revealed chloroma. MRI is the method that best characterizes this lesion. (orig.). With 5 figs.

  15. Hyperspectral Image Classification Based on the Combination of Spatial-spectral Feature and Sparse Representation

    YANG Zhaoxia


    Full Text Available In order to avoid the problem of being over-dependent on high-dimensional spectral feature in the traditional hyperspectral image classification, a novel approach based on the combination of spatial-spectral feature and sparse representation is proposed in this paper. Firstly, we extract the spatial-spectral feature by reorganizing the local image patch with the first d principal components(PCs into a vector representation, followed by a sorting scheme to make the vector invariant to local image rotation. Secondly, we learn the dictionary through a supervised method, and use it to code the features from test samples afterwards. Finally, we embed the resulting sparse feature coding into the support vector machine(SVM for hyperspectral image classification. Experiments using three hyperspectral data show that the proposed method can effectively improve the classification accuracy comparing with traditional classification methods.

  16. MR imaging features of foot involvement in ankylosing spondylitis

    Erdem, C. Zuhal E-mail:; Sarikaya, Selda; Erdem, L. Oktay; Ozdolap, Senay; Gundogdu, Sadi


    Objective: To determine alterations of the soft tissue, tendon, cartilage, joint space, and bone of the foot using magnetic resonance (MR) imaging in ankylosing spondylitis (AS) patients. Materials and Method: Clinical and MR examination of the foot was performed in 23 AS patients (46 feet). Ten asymptomatic volunteers (20 feet) were studied on MR imaging, as a control group. MR imaging protocol included; T1-weighted spin-echo, T2-weighted fast-field echo (FFE) and fat-suppressed short tau inversion recovery (STIR) sequences in sagittal, sagittal oblique, and coronal planes using a head coil. Specifically, we examined: bone erosions, tendinitis (acute and chronic), para-articular enthesophyte, joint effusion, plantar fasciitis, joint space narrowing, soft tissue edema, bone marrow edema, enthesopathy in the Achilles tendon and plantar fascia attachment, subchondral signal intensity abnormalities (edema and sclerosis), tenosynovitis, retrocalcaneal bursitis, subchondral cysts, subchondral fissures, and bony ankylosis. Midfoot, hindfoot, and ankle were included in examined anatomic regions. Results: Clinical signs and symptoms (pain and swelling) due to foot involvement were present in 3 (13%) of the patients while frequency of involvement was 21 (91%) with MR imaging assessment. The MR imaging findings were bone erosions (65%), Achilles tendinitis (acute and chronic) (61%), para-articular enthesophyte (48%), joint effusion (43%), plantar fasciitis (40%), joint space narrowing (40%), subchondral sclerosis (35%), soft tissue edema (30%), bone marrow edema (30%), enthesopathy of the Achilles attachment (30%), subchondral edema (26%), enthesopathy in the plantar fascia attachment (22%), retrocalcaneal bursitis (22%), subchondral cysts (17%), subchondral fissures (17%), tendinitis and enthesopathy of the plantar ligament (13%), and bony ankylosis (9%). The most common involved anatomical region was the hindfoot (83%) following by midfoot (69% ) and ankle (22

  17. Content-Based Image Retrieval using Color Moment and Gabor Texture Feature

    K. Hemachandran


    Full Text Available Content based image retrieval (CBIR has become one of the most active research areas in the past few years. Many indexing techniques are based on global feature distributions. However, these global distributions have limited discriminating power because they are unable to capture local image information. In this paper, we propose a content-based image retrieval method which combines color and texture features. To improve the discriminating power of color indexing techniques, we encode a minimal amount of spatial information in the color index. As its color features, an image is divided horizontally into three equal non-overlapping regions. From each region in the image, we extract the first three moments of the color distribution, from each color channel and store them in the index i.e., for a HSV color space, we store 27 floating point numbers per image. As its texture feature, Gabor texture descriptors are adopted. We assign weights to each feature respectively and calculate the similarity with combined features of color and texture using Canberra distance as similarity measure. Experimental results show that the proposed method has higher retrieval accuracy than other conventional methods combining color moments and texture features based on global features approach.

  18. Second order Statistical Texture Features from a New CSLBPGLCM for Ultrasound Kidney Images Retrieval



    Full Text Available This work proposes a new method called Center Symmetric Local Binary Pattern Grey Level Co-occurrence Matrix (CSLBPGLCM for the purpose of extracting second order statistical texture features in ultrasound kidney images. These features are then feed into ultrasound kidney images retrieval system for the point of medical applications. This new GLCM matrix combines the benefit of CSLBP and conventional GLCM. The main intention of this CSLBPGLCM is to reduce the number of grey levels in an image by not simply accumulating the grey levels but incorporating another statistical texture feature in it. The proposed approach is cautiously evaluated in ultrasound kidney images retrieval system and has been compared with conventional GLCM. It is experimentally proved that the proposed method increases the retrieval efficiency, accuracy and reduces the time complexity of ultrasound kidney images retrieval system by means of second order statistical texture features.

  19. Imaging features of gossypiboma: report of two cases.

    Prasad S


    Full Text Available Recognition of postoperatively retained foreign body referred euphemistically as gossypiboma is essential but is very often considerably delayed. Legal implications as well as confusing configuration patterns cause considerable dilemma in the accurate diagnosis. We present computed tomographic features of gossypiboma in two patients who presented with symptoms of fever and pain in the immediate post-operative period. A prospective radiological diagnosis is essential for further management in these patients.

  20. Simultaneous Spectral-Spatial Feature Selection and Extraction for Hyperspectral Images.

    Zhang, Lefei; Zhang, Qian; Du, Bo; Huang, Xin; Tang, Yuan Yan; Tao, Dacheng


    In hyperspectral remote sensing data mining, it is important to take into account of both spectral and spatial information, such as the spectral signature, texture feature, and morphological property, to improve the performances, e.g., the image classification accuracy. In a feature representation point of view, a nature approach to handle this situation is to concatenate the spectral and spatial features into a single but high dimensional vector and then apply a certain dimension reduction technique directly on that concatenated vector before feed it into the subsequent classifier. However, multiple features from various domains definitely have different physical meanings and statistical properties, and thus such concatenation has not efficiently explore the complementary properties among different features, which should benefit for boost the feature discriminability. Furthermore, it is also difficult to interpret the transformed results of the concatenated vector. Consequently, finding a physically meaningful consensus low dimensional feature representation of original multiple features is still a challenging task. In order to address these issues, we propose a novel feature learning framework, i.e., the simultaneous spectral-spatial feature selection and extraction algorithm, for hyperspectral images spectral-spatial feature representation and classification. Specifically, the proposed method learns a latent low dimensional subspace by projecting the spectral-spatial feature into a common feature space, where the complementary information has been effectively exploited, and simultaneously, only the most significant original features have been transformed. Encouraging experimental results on three public available hyperspectral remote sensing datasets confirm that our proposed method is effective and efficient.

  1. Large-scale computations on histology images reveal grade-differentiating parameters for breast cancer

    Katsinis Constantine


    Full Text Available Abstract Background Tumor classification is inexact and largely dependent on the qualitative pathological examination of the images of the tumor tissue slides. In this study, our aim was to develop an automated computational method to classify Hematoxylin and Eosin (H&E stained tissue sections based on cancer tissue texture features. Methods Image processing of histology slide images was used to detect and identify adipose tissue, extracellular matrix, morphologically distinct cell nuclei types, and the tubular architecture. The texture parameters derived from image analysis were then applied to classify images in a supervised classification scheme using histologic grade of a testing set as guidance. Results The histologic grade assigned by pathologists to invasive breast carcinoma images strongly correlated with both the presence and extent of cell nuclei with dispersed chromatin and the architecture, specifically the extent of presence of tubular cross sections. The two parameters that differentiated tumor grade found in this study were (1 the number density of cell nuclei with dispersed chromatin and (2 the number density of tubular cross sections identified through image processing as white blobs that were surrounded by a continuous string of cell nuclei. Classification based on subdivisions of a whole slide image containing a high concentration of cancer cell nuclei consistently agreed with the grade classification of the entire slide. Conclusion The automated image analysis and classification presented in this study demonstrate the feasibility of developing clinically relevant classification of histology images based on micro- texture. This method provides pathologists an invaluable quantitative tool for evaluation of the components of the Nottingham system for breast tumor grading and avoid intra-observer variability thus increasing the consistency of the decision-making process.

  2. Melancholic features in inpatients with major depressive disorder associate with differential clinical characteristics and treatment outcomes.

    Lin, Ching-Hua; Huang, Chun-Jen; Liu, Shi-Kai


    To determine whether the presence of melancholic features in hospitalized patients with major depressive disorder (MDD) was associated with specific clinical characteristics and treatment outcomes, supporting melancholic depression as a distinct subtype within MDD. 126 acutely ill inpatients with MDD were enrolled in an open, 6-week trial with fixed-dose fluoxetine 20mg daily. Symptom severity was assessed regularly, using the 17-item Hamilton Depression Rating Scale (HAMD-17) and Clinical Global Impression of Severity (CGI-S). Melancholic features were defined according to the DSM-IV criteria. Clinical variables were compared between patients with and without melancholic features. Generalized estimating equations method was used to explore the differences in HAMD-17 and CGI-S scores between the 2 groups over time. Clinical response was defined as having a 50% or greater reduction in HAMD-17 scores. 96 (76.2%) of the 126 patients with at least one post-baseline assessment met the criteria for melancholic depression. Melancholic depression differed from non-melancholic depression in clinical characteristics and predicted a better response to fluoxetine treatment. The differentiation between melancholic and non-melancholic depression within MDD hence is clinically significant and valid. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.

  3. Texture Feature Analysis for Different Resolution Level of Kidney Ultrasound Images

    Kairuddin, Wan Nur Hafsha Wan; Mahmud, Wan Mahani Hafizah Wan


    Image feature extraction is a technique to identify the characteristic of the image. The objective of this work is to discover the texture features that best describe a tissue characteristic of a healthy kidney from ultrasound (US) image. Three ultrasound machines that have different specifications are used in order to get a different quality (different resolution) of the image. Initially, the acquired images are pre-processed to de-noise the speckle to ensure the image preserve the pixels in a region of interest (ROI) for further extraction. Gaussian Low- pass Filter is chosen as the filtering method in this work. 150 of enhanced images then are segmented by creating a foreground and background of image where the mask is created to eliminate some unwanted intensity values. Statistical based texture features method is used namely Intensity Histogram (IH), Gray-Level Co-Occurance Matrix (GLCM) and Gray-level run-length matrix (GLRLM).This method is depends on the spatial distribution of intensity values or gray levels in the kidney region. By using One-Way ANOVA in SPSS, the result indicated that three features (Contrast, Difference Variance and Inverse Difference Moment Normalized) from GLCM are not statistically significant; this concludes that these three features describe a healthy kidney characteristics regardless of the ultrasound image quality.

  4. Spinal dural arteriovenous fistula: Imaging features and its mimics

    Jeog, Ying; Ting, David Yen; Hsu, Hui Ling; Huang, Yen Lin; Chen, Chi Jen; Tseng, Ting Chi [Dept. of Radiology, aipei Medical University-Shuang Ho Hospital, New Taipei City, Taiwan (China)


    Spinal dural arteriovenous fistula (SDAVF) is the most common spinal vascular malformation, however it is still rare and underdiagnosed. Magnetic resonance imaging findings such as spinal cord edema and dilated and tortuous perimedullary veins play a pivotal role in the confirmation of the diagnosis. However, spinal angiography remains the gold standard in the diagnosis of SDAVF. Classic angiographic findings of SDAVF are early filling of radicular veins, delayed venous return, and an extensive network of dilated perimedullary venous plexus. A series of angiograms of SDAVF at different locations along the spinal column, and mimics of serpentine perimedullary venous plexus on MR images, are demonstrated. Thorough knowledge of SDAVF aids correct diagnosis and prevents irreversible complications.

  5. Perfusion MR imaging for differentiation of benign and malignant meningiomas

    Zhang, Hao [University of Groningen, Department of Radiology, University Medical Center Groningen, Groningen (Netherlands); Shanghai Jiaotong University, Department of Radiology, First People' s Hospital, Shanghai (China); Roediger, Lars A.; Oudkerk, Matthijs [University of Groningen, Department of Radiology, University Medical Center Groningen, Groningen (Netherlands); Shen, Tianzhen [Fudan University, Department of Radiology, Huashan Hospital, Shanghai (China); Miao, Jingtao [Shanghai Jiaotong University, Department of Radiology, First People' s Hospital, Shanghai (China)


    Our purpose was to determine whether perfusion MR imaging can be used to differentiate benign and malignant meningiomas on the basis of the differences in perfusion of tumor parenchyma and/or peritumoral edema. A total of 33 patients with preoperative meningiomas (25 benign and 8 malignant) underwent conventional and dynamic susceptibility contrast perfusion MR imaging. Maximal relative cerebral blood volume (rCBV) and the corresponding relative mean time to enhance (rMTE) (relative to the contralateral normal white matter) in both tumor parenchyma and peritumoral edema were measured. The independent samples t-test was used to determine whether there was a statistically significant difference in the mean rCBV and rMTE ratios between benign and malignant meningiomas. The mean maximal rCBV values of benign and malignant meningiomas were 7.16{+-}4.08 (mean{+-}SD) and 5.89{+-}3.86, respectively, in the parenchyma, and 1.05{+-}0.96 and 3.82{+-}1.39, respectively, in the peritumoral edema. The mean rMTE values were 1.16{+-}0.24 and 1.30{+-}0.32, respectively, in the parenchyma, and 0.91{+-}0.25 and 1.24{+-}0.35, respectively, in the peritumoral edema. The differences in rCBV and rMTE values between benign and malignant meningiomas were not statistically significant (P>0.05) in the parenchyma, but both were statistically significant (P<0.05) in the peritumoral edema. Perfusion MR imaging can provide useful information on meningioma vascularity which is not available from conventional MRI. Measurement of maximal rCBV and corresponding rMTE values in the peritumoral edema is useful in the preoperative differentiation between benign and malignant meningiomas. (orig.)

  6. Malignant Peripheral Nerve Sheath Tumors: Differentiation Patterns and Immunohistochemical Features - A Mini-Review and Our New Findings

    Guo, Aitao; Liu,Aijun; Wei, Lixin; Song, Xin


    Malignant peripheral nerve sheath tumors (MPNST) represent a group of highly heterogeneous human malignancies often with multiple histological origins, divergent differentiation patterns, and diverse immunohistochemical presentations. The differential diagnosis of MPNST from other spindle cell neoplasms poses great challenges for pathologists. This report provides a mini-review of these unique features associated with MPNST and also presents the first cases of MPNST with six differentiation p...

  7. Malignant Peripheral Nerve Sheath Tumors: Differentiation Patterns and Immunohistochemical Features - A Mini-Review and Our New Findings

    Aitao Guo, Aijun Liu, Lixin Wei, Xin Song


    Full Text Available Malignant peripheral nerve sheath tumors (MPNST represent a group of highly heterogeneous human malignancies often with multiple histological origins, divergent differentiation patterns, and diverse immunohistochemical presentations. The differential diagnosis of MPNST from other spindle cell neoplasms poses great challenges for pathologists. This report provides a mini-review of these unique features associated with MPNST and also presents the first cases of MPNST with six differentiation patterns.

  8. Edge-Based Feature Extraction Method and Its Application to Image Retrieval

    G. Ohashi


    Full Text Available We propose a novel feature extraction method for content-bases image retrieval using graphical rough sketches. The proposed method extracts features based on the shape and texture of objects. This edge-based feature extraction method functions by representing the relative positional relationship between edge pixels, and has the advantage of being shift-, scale-, and rotation-invariant. In order to verify its effectiveness, we applied the proposed method to 1,650 images obtained from the Hamamatsu-city Museum of Musical Instruments and 5,500 images obtained from Corel Photo Gallery. The results verified that the proposed method is an effective tool for achieving accurate retrieval.

  9. Two-dimensional electrophoresis analysis of proteomics based on image feature and mathematical morphology

    SHEN Peng; FAN Xiaohui; ZENG Zhen; CHENG Yiyu


    In this paper, a novel method to automatically detect protein spots on a two-dimensional (2-D) electrophoresis gel image is proposed to implement proteomics analysis of complex analyte.On the basis of the identifying spots results based on color variation and spot size features, morphological feature is introduced as a new criterion to distinguish protein spots from non-protein spots.Image-sharpening, edge-detecting and morphological feature extraction methods were consequently combined to detect protein spots on a 2-D electrophoresis gel image subject to strong disturbance.The proposed method was applied to detect the protein spots of proteomic gel images from E.coli cell, human kidney tissue and human serum.The results demonstrated that this method is more accurate and reliable than previous methods such as PDQuest 7.2 and ImageMaster 5.0 software for detecting protein spots on gel images with strong interferences.

  10. Comparisons of feature extraction algorithm based on unmanned aerial vehicle image

    Xi, Wenfei; Shi, Zhengtao; Li, Dongsheng


    Feature point extraction technology has become a research hotspot in the photogrammetry and computer vision. The commonly used point feature extraction operators are SIFT operator, Forstner operator, Harris operator and Moravec operator, etc. With the high spatial resolution characteristics, UAV image is different from the traditional aviation image. Based on these characteristics of the unmanned aerial vehicle (UAV), this paper uses several operators referred above to extract feature points from the building images, grassland images, shrubbery images, and vegetable greenhouses images. Through the practical case analysis, the performance, advantages, disadvantages and adaptability of each algorithm are compared and analyzed by considering their speed and accuracy. Finally, the suggestions of how to adapt different algorithms in diverse environment are proposed.

  11. Featured Image: Violent History of the Toothbrush Cluster

    Kohler, Susanna


    This stunning composite image shows the components of the galaxy cluster RX J0603.3+4214, located at a redshift of z=0.225. This image contains Chandra X-ray data (red), radio data from the Giant Metrewave Radio Telescope (green), and optical from the Subaru Telescope (background). The shape of the enormous (6.5 million light-years across!) radio relic, shown in green, gives this collection of galaxies its nickname: the Toothbrush Cluster. A team of scientists led by Myungkook James Jee (Yonsei University and University of California, Davis) used Hubble and Subaru to study weak gravitational lensing by the Toothbrush Cluster, in order to determine how the clusters mass is distributed. Jee and collaborators found that most of the dark-matter mass is located in two large clumps on a north-south axis (shown by the white contours overlaid on the image), suggesting that the Toothbrush Cluster is the result of a past merger between two clusters. This violent merger is likely what caused the enormous Toothbrush radio relic. Check out the paper below for more information!CitationM. James Jee et al 2016 ApJ 817 179. doi:10.3847/0004-637X/817/2/179

  12. Hepatic CT Image Query Based on Threshold-based Classification Scheme with Gabor Features

    JIANG Li-jun; LUO Yong-zing; ZHAO Jun; ZHUANG Tian-ge


    Hepatic computed tomography (CT) images with Gabor function were analyzed.Then a thresholdbased classification scheme was proposed using Gabor features and proceeded with the retrieval of the hepatic CT images.In our experiments,a batch of hepatic CT images containing several types of CT findings was used and compared with the Zhao's image classification scheme,support vector machines (SVM) scheme and threshold-based scheme.

  13. Prediction of neural differentiation fate of rat mesenchymal stem cells by quantitative morphological analyses using image processing techniques.

    Kazemimoghadam, Mahdieh; Janmaleki, Mohsen; Fouani, Mohamad Hassan; Abbasi, Sara


    Differentiation of bone marrow mesenchymal stem cells (BMSCs) into neural cells has received significant attention in recent years. However, there is still no practical method to evaluate differentiation process non-invasively and practically. The cellular quality evaluation method is still limited to conventional techniques, which are based on extracting genes or proteins from the cells. These techniques are invasive, costly, time consuming, and should be performed by relevant experts in equipped laboratories. Moreover, they cannot anticipate the future status of cells. Recently, cell morphology has been introduced as a feasible way of monitoring cell behavior because of its relationship with cell proliferation, functions and differentiation. In this study, rat BMSCs were induced to differentiate into neurons. Subsequently, phase contrast images of cells taken at certain intervals were subjected to a series of image processing steps and cell morphology features were calculated. In order to validate the viability of applying image-based approaches for estimating the quality of differentiation process, neural-specific markers were measured experimentally throughout the induction. The strong correlation between quantitative imaging metrics and experimental outcomes revealed the capability of the proposed approach as an auxiliary method of assessing cell behavior during differentiation.

  14. Morphologic patterns and imaging features of intracranial hemangiopericytomas: a retrospective analysis

    Pang HP


    Full Text Available Haopeng Pang,1 Zhenwei Yao,1 Yan Ren,1 Guobing Liu,2 Jiawen Zhang,1 Xiaoyuan Feng1 1Department of Radiology, Affiliated Huashan Hospital of Fudan University, People’s Republic of China; 2Department of Nuclear Medicine, Zhongshan Hospital, Fudan University, Shanghai, People’s Republic of China Objectives: Hemangiopericytomas (HPCs are rare intracranial tumors. Their differential diagnosis using computed tomography (CT and magnetic resonance imaging (MRI is difficult because of similarities in morphologic features with other intracranial tumors and meningiomas. Methods: We retrospectively analyzed the clinical data and CT and MRI findings of 32 patients diagnosed with HPCs via histopathology. We evaluated the location, shape, morphologic patterns, density, and signal intensity of the tumors and classified them into four types. Results: The number of tumors analyzed was 32; 29 were supratentorial and three were infratentorial. Eighteen tumors were lobular, while 14 were oval in shape. Further, 28 tumors had cystic areas, and 16 had signal-void vessels. Among the 20 tumors that had been scanned by MRI; eleven showed isointensity, eight slight hyperintensity, and one slight hypointensity on T1-weighted image. Moreover, 12 showed isointensity, and eight showed slight hyperintensity on T2-weighted image and T2-weighted-fluid-attenuated-inversion recovery. Diffusion-weighted images showed isointensity (9/13 or slight hyperintensity (4/13. Of the 15 tumors scanned by contrast-enhanced MRI, one showed poor enhancement; six, moderate enhancement; and eight, intense enhancement. Only one tumor exhibited the “dural tail” sign. Moreover, calcification was observed in just one tumor on CT imaging (1/22. All tumors (5/5 showed intense enhancement on CT angiography, whereas some exhibited dual blood supply (2/5. Conclusion: We conclude that tumors present outside the brain parenchyma, with isointense to slightly intense regions on MRI scans, oval

  15. [Classification technique for hyperspectral image based on subspace of bands feature extraction and LS-SVM].

    Gao, Heng-zhen; Wan, Jian-wei; Zhu, Zhen-zhen; Wang, Li-bao; Nian, Yong-jian


    The present paper proposes a novel hyperspectral image classification algorithm based on LS-SVM (least squares support vector machine). The LS-SVM uses the features extracted from subspace of bands (SOB). The maximum noise fraction (MNF) method is adopted as the feature extraction method. The spectral correlations of the hyperspectral image are used in order to divide the feature space into several SOBs. Then the MNF is used to extract characteristic features of the SOBs. The extracted features are combined into the feature vector for classification. So the strong bands correlation is avoided and the spectral redundancies are reduced. The LS-SVM classifier is adopted, which replaces inequality constraints in SVM by equality constraints. So the computation consumption is reduced and the learning performance is improved. The proposed method optimizes spectral information by feature extraction and reduces the spectral noise. The classifier performance is improved. Experimental results show the superiorities of the proposed algorithm.

  16. A Content based CT Lung Image Retrieval by DCT Matrix and Feature Vector Technique

    J.Bridget Nirmala


    Full Text Available Most of the image retrieval systems are still incapable of providing retrieval result with high retrieval accuracy and less computational complexity. Image Retrieval technique to retrieve similar and relevant Computed Tomography (CT images of lung from a large database of images. During the process of retrieval, a query image which contains the affected area / abnormal region is given as an input to retrieve similar images which contain affected area/abnormal region from the database. DCT Matrix (DCTM is a kind of commonly used color feature representation in image retrieval. This paper describes a content based image retrieval (CBIR that represent each image in database by a vector of feature values called DCT vector matrix(8x8. Using this DCTM row and column feature vector values considered as a query image which is compared with existing database to cull out more similar and relevant images. The experimental result shows that 97% of images can be retrieved correctly using this technique

  17. A PCA Based Automatic Image Categorization Approach Using Dominant Color Features

    WUChunming; QIANHui; WANGDonghui


    Automatic Image categorization is a universal problem in area of Content-based image retrieval (CBIR). The goal of automatic image categorization is to find a mapping between images and the predefined image categories. The difficulty of this problem is that how to describe image content and incorporate low-level features into semantic categories. As a solution, we propose a Principal component analysis (PCA) based approach. This approach assumes that the images in the same semantic category have the similar spatial distribution of color components and treats the images in the same category as a linear combination of a fixed set of dominant color blocks with special textural information. A three-step algorithm is designed: (1) extracting Dominant colors (DC) of images, which describe the major color information in an image; (2) Establishing a feature space based on DC blocks and its textural information; (3) using PCA to reduce dimensionality of feature space and using the basis vectors to categorize images. An experimental database containing nine categories including cars, flowers, houses, portraits, fish, bark, sunshine, leaves and fresco is constructed to test the algorithm based on our image categorization approach. The results show that this approach is effective and a reasonable compromise between accuracy and speed in practice.

  18. Differential diagnosis of ACTH-dependent hypercortisolism: imaging versus laboratory.

    Andrioli, Massimiliano; Pecori Giraldi, Francesca; De Martin, Martina; Cattaneo, Agnese; Carzaniga, Chiara; Cavagnini, Francesco


    Differential diagnosis of ACTH-dependent Cushing's syndrome often presents major difficulties. Diagnostic troubles are increased by suboptimal specificity of endocrine tests, the rarity of ectopic ACTH secretion and the frequent incidental discovery of pituitary adenomas. A 43-year-old female reported with mild signs and symptoms of hypercortisolism, and initial hormonal tests and results of pituitary imaging (7-mm adenoma) were suggestive for Cushing's disease. However, inadequate response to corticotrophin-releasing hormone and failure to suppress after 8 mg dexamethasone pointed towards an ectopic source. Total body CT scan visualized only a small, non-specific nodule in the right posterior costophrenic excavation. Inferior petrosal sinus sampling revealed an absent center:periphery ACTH gradient but octreoscan and (18)F-FDG-PET-CT failed to detect abnormal tracer accumulation. We weighed results of the laboratory with those of imaging and decided to remove the lung nodule. Pathology identified a typical, ACTH-staining carcinoid and the diagnosis was confirmed by postsurgical hypoadrenalism. In conclusion, imaging may prove unsatisfactory or even misleading for the etiologial diagnosis of ACTH-dependent Cushing's syndrome and should therefore be interpreted only in context with results of hormonal dynamic testing.

  19. Spinal cord cavities; Differential-diagnostic criteria in magnetic resonance imaging

    Schubeus, P.; Schoerner, W.; Hosten, N.; Felix, R. (Free University of Berlin, University Clinic Rudolf Virchow, Charlottenburg (Germany). Department of Radiology)

    MRI examinations of 30 patients with idiopathic syringomyelia and 10 patients with cavities associated with an intramedullary neoplasm were evaluated with respect to typical MRI features in both groups. Al tumor-associated cases resembled the idiopathic syringomyelias in some portions of the cavity. At the tumor site, however, tumor-associated cases demonstrated typical findings; the cavities showed abrupt changes of diameter (10/10) and position (8/10) and the surrounding spinal cord demonstrated an uneven thickness (10/10), an increased signal intensity on T2-weighted images (10/10) and pathological contrast enhancement (7/7). Displacement of cerebellar tonsils below the level of the foramen magnum (921/30) and enlargement of the spinal canal (97/29) were characteristic features of idiopathic cases. In conclusion, MRI provides valuable criteria to differentiate between idiopathic and tumor-associated cavities. (author). 19 refs.; 4 figs.; 1 tab.

  20. Feature learning and change feature classification based on deep learning for ternary change detection in SAR images

    Gong, Maoguo; Yang, Hailun; Zhang, Puzhao


    Ternary change detection aims to detect changes and group the changes into positive change and negative change. It is of great significance in the joint interpretation of spatial-temporal synthetic aperture radar images. In this study, sparse autoencoder, convolutional neural networks (CNN) and unsupervised clustering are combined to solve ternary change detection problem without any supervison. Firstly, sparse autoencoder is used to transform log-ratio difference image into a suitable feature space for extracting key changes and suppressing outliers and noise. And then the learned features are clustered into three classes, which are taken as the pseudo labels for training a CNN model as change feature classifier. The reliable training samples for CNN are selected from the feature maps learned by sparse autoencoder with certain selection rules. Having training samples and the corresponding pseudo labels, the CNN model can be trained by using back propagation with stochastic gradient descent. During its training procedure, CNN is driven to learn the concept of change, and more powerful model is established to distinguish different types of changes. Unlike the traditional methods, the proposed framework integrates the merits of sparse autoencoder and CNN to learn more robust difference representations and the concept of change for ternary change detection. Experimental results on real datasets validate the effectiveness and superiority of the proposed framework.