Gray-level Features Research Articles

Segmentation of retinal blood vessels in colour fundus images using ANFIS classifier

Diabetic Retinopathy is an irreversible retinal disorder in the diabetic patients where the blood vessels are injured due to the hyper pressure or flow of the blood through the vessels in retina. The blood vessels originate from the centre of the optic disc and spreads over the entire region of retina. Vision loss in diabetic patients can be prevented at an earlier stage if blood vessels are screened initially. Hence, detection of retinal blood vessels is significant for DR detection. In this paper, the computer-aided automatic detection and segmentation of retinal blood vessel is proposed by extracting the features based on orientation analysis of gradient vector field, morphological and gray level and Gabor features and are then classified using ANFIS classifier. The results show that the proposed method achieves sensitivity of 92.66%, 81.23% specificity of 98.53%, 98.70% and accuracy of 98.45%, 95.20% on DRIVE and STARE dataset.

Copy-Move forgery detection in images using grey-level run length matrix features

Copy-move detection also known as region duplication detection is a well-recognised and active area of research owing to great demand of authenticating genuineness of images. Currently available techniques for copy-move detection fail to accurately locate the tampered region and lack robustness against common post-processing operations like compression, blurring and brightness changes. This paper proposes a novel technique for detection and localisation of copy-move regions in image using grey-level run length matrix (GLRLM) features. In the proposed method, we first divide the forged image into overlapping blocks and GLRLM features are calculated for each block. Features calculated for each block form feature vectors. Feature vectors thus obtained are lexicographically sorted. Blocks with similar features are identified using Euclidean feature distances. Post-processing isolates similar blocks. The results demonstrate the effectiveness of the proposed scheme to locate copy-move forgery and robustness against operations like jpeg compression, blurring and contrast adjustments.

Mass detection in digital breast tomosynthesis: Deep convolutional neural network with transfer learning from mammography.

Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms. A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a linear discriminant classifier to score the detected masses. For the DCNN-based CAD system, ROIs from five consecutive slices centered at each candidate were passed through the trained DCNN and a mass likelihood score was generated. The performances of the CAD systems were evaluated using free-response ROC curves and the performance difference was analyzed using a non-parametric method. Before transfer learning, the DCNN trained only on mammograms with an AUC of 0.99 classified DBT masses with an AUC of 0.81 in the DBT training set. After transfer learning with DBT, the AUC improved to 0.90. For breast-based CAD detection in the test set, the sensitivity for the feature-based and the DCNN-based CAD systems was 83% and 91%, respectively, at 1 FP/DBT volume. The difference between the performances for the two systems was statistically significant (p-value < 0.05). The image patterns learned from the mammograms were transferred to the mass detection on DBT slices through the DCNN. This study demonstrated that large data sets collected from mammography are useful for developing new CAD systems for DBT, alleviating the problem and effort of collecting entirely new large data sets for the new modality.

Method of Image De-Noising Based on Non-Noisy Atoms Self Adaptive and Sparse Representation

In allusion to the losses of image detail and texture structure information during image de-noising process, an image de-noising algorithm based on non-related dictionary learning is proposed in this paper. Firstly, this algorithm is adopted to obtain selfadaption redundant dictionary for the noisy image through the dictionary learning algorithm; secondly, HOG features and gray-level statistical features of each atom in the dictionary are extracted to form the feature set, and meanwhile the feature set of the atoms is adopted to divide the atoms into two types (non-noisy atoms and noisy atoms); finally, the non-noisy atoms are adopted to recover the image, thus to realize the denosing purpose. The experiment result shows: the proposed algorithm does not need to know the prior information of the noise and PSNR performance thereof is better than that of existing algorithms, and meanwhile the proposed algorithm can well keep the image detail and texture structure information, thus to improve visual effect.

Fast retrieval of calcification from sequential intravascular ultrasound gray-scale images.

Intravascular ultrasound (IVUS)-based tissue characterization is invaluable for the computer-aided diagnosis and interventional treatment of cardiac vessel diseases. Although the analysis of raw backscattered signals allows more accurate plaque characterization than gray-scale images, its applications are limited due to its nature of electrocardiogram-gated acquisition. Images acquired by IVUS devices that do not allow the acquisition of raw signals cannot be characterized. To address these limitations, we developed a method for fast frame-by-frame retrieval and location of calcification according to the jump features of radial gray-level variation curves from sequential IVUS gray-scale images. The proposed method consists of three main steps: (1) radial gray-level variation curves are extracted from each filtered polar view, (2) sequential images are preliminarily queried according to the maximal slopes of radial gray-level variation curves, and finally, (3) key frames that include calcification are selected through checking the gray-level features of successive pixel columns in the preliminary results. Experimental results with clinically acquired in vivo data sets indicate key frames that include calcification can be retrieved with the advantages of simplicity, high efficiency, and accuracy. Recognition results correlate well with manual characterization results obtained by experienced physicians and through virtual histology.

Side scan sonar image segmentation based on neutrosophic set and quantum-behaved particle swarm optimization algorithm

To fulfill side scan sonar (SSS) image segmentation accurately and efficiently, a novel segmentation algorithm based on neutrosophic set (NS) and quantum-behaved particle swarm optimization (QPSO) is proposed in this paper. Firstly, the neutrosophic subset images are obtained by transforming the input image into the NS domain. Then, a co-occurrence matrix is accurately constructed based on these subset images, and the entropy of the gray level image is described to serve as the fitness function of the QPSO algorithm. Moreover, the optimal two-dimensional segmentation threshold vector is quickly obtained by QPSO. Finally, the contours of the interested target are segmented with the threshold vector and extracted by the mathematic morphology operation. To further improve the segmentation efficiency, the single threshold segmentation, an alternative algorithm, is recommended for the shadow segmentation by considering the gray level characteristics of the shadow. The accuracy and efficiency of the proposed algorithm are assessed with experiments of SSS image segmentation.

Fission gas bubble identification using MATLAB's image processing toolbox

Automated image processing routines have the potential to aid in the fuel performance evaluation process by eliminating bias in human judgment that may vary from person-to-person or sample-to-sample. This study presents several MATLAB based image analysis routines designed for fission gas void identification in post-irradiation examination of uranium molybdenum (U–Mo) monolithic-type plate fuels. Frequency domain filtration, enlisted as a pre-processing technique, can eliminate artifacts from the image without compromising the critical features of interest. This process is coupled with a bilateral filter, an edge-preserving noise removal technique aimed at preparing the image for optimal segmentation. Adaptive thresholding proved to be the most consistent gray-level feature segmentation technique for U–Mo fuel microstructures. The Sauvola adaptive threshold technique segments the image based on histogram weighting factors in stable contrast regions and local statistics in variable contrast regions. Once all processing is complete, the algorithm outputs the total fission gas void count, the mean void size, and the average porosity. The final results demonstrate an ability to extract fission gas void morphological data faster, more consistently, and at least as accurately as manual segmentation methods.

SU-F-R-31: Identification of Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induced Lung Disease

Purpose: Normal lung CT texture features have been used for the prediction of radiation-induced lung disease (radiation pneumonitis and radiation fibrosis). For these features to be clinically useful, they need to be relatively invariant (robust) to tumor size and not correlated with normal lung volume. Methods: The free-breathing CTs of 14 lung SBRT patients were studied. Different sizes of GTVs were simulated with spheres placed at the upper lobe and lower lobe respectively in the normal lung (contralateral to tumor). 27 texture features (9 from intensity histogram, 8 from grey-level co-occurrence matrix [GLCM] and 10 from grey-level run-length matrix [GLRM]) were extracted from [normal lung-GTV]. To measure the variability of a feature F, the relative difference D=|Fref -Fsim|/Fref*100% was calculated, where Fref was for the entire normal lung and Fsim was for [normal lung-GTV]. A feature was considered as robust if the largest non-outlier (Q3+1.5*IQR) D was less than 5%, and considered as not correlated with normal lung volume when their Pearson correlation was lower than 0.50. Results: Only 11 features were robust. All first-order intensity-histogram features (mean, max, etc.) were robust, while most higher-order features (skewness, kurtosis, etc.) were unrobust. Only two of the GLCM and four of the GLRM features were robust. Larger GTV resulted greater feature variation, this was particularly true for unrobust features. All robust features were not correlated with normal lung volume while three unrobust features showed high correlation. Excessive variations were observed in two low grey-level run features and were later identified to be from one patient with local lung diseases (atelectasis) in the normal lung. There was no dependence on GTV location. Conclusion: We identified 11 robust normal lung CT texture features that can be further examined for the prediction of radiation-induced lung disease. Interestingly, low grey-level run features identified normal lung diseases. This work was supported in part by the National Cancer Institute Grants R01CA172638.

Method for simulating dose reduction in digital mammography using the Anscombe transformation.

Purpose:This work proposes an accurate method for simulating dose reduction in digital mammography starting from a clinical image acquired with a standard dose.Methods:The method developed in this work consists of scaling a mammogram acquired at the standard radiation dose and adding signal-dependent noise. The algorithm accounts for specific issues relevant in digital mammography images, such as anisotropic noise, spatial variations in pixel gain, and the effect of dose reduction on the detective quantum efficiency. The scaling process takes into account the linearity of the system and the offset of the detector elements. The inserted noise is obtained by acquiring images of a flat-field phantom at the standard radiation dose and at the simulated dose. Using the Anscombe transformation, a relationship is created between the calculated noise mask and the scaled image, resulting in a clinical mammogram with the same noise and gray level characteristics as an image acquired at the lower-radiation dose.Results:The performance of the proposed algorithm was validated using real images acquired with an anthropomorphic breast phantom at four different doses, with five exposures for each dose and 256 nonoverlapping ROIs extracted from each image and with uniform images. The authors simulated lower-dose images and compared these with the real images. The authors evaluated the similarity between the normalized noise power spectrum (NNPS) and power spectrum (PS) of simulated images and real images acquired with the same dose. The maximum relative error was less than 2.5% for every ROI. The added noise was also evaluated by measuring the local variance in the real and simulated images. The relative average error for the local variance was smaller than 1%.Conclusions:A new method is proposed for simulating dose reduction in clinical mammograms. In this method, the dependency between image noise and image signal is addressed using a novel application of the Anscombe transformation. NNPS, PS, and local noise metrics confirm that this method is capable of precisely simulating various dose reductions.

Abstract P4-01-02: Identification of histopathologic determinants of mammographic breast density as a cancer risk factor

Abstract Background: Mammographic density and texture, the amount and appearance of fibroglandular tissue respectively, are known to be strong predictors of a woman's risk to develop breast cancer. However, how differences in underlying tissue biology are associated with the wide range of breast density patterns seen among women is currently unknown. Understanding the biological mechanisms driving breast density can be critical for developing targeted interventions aimed at reducing a woman's risk for breast cancer. Methods: Digital mammograms and formalin-fixed paraffin-embedded biopsy tissue sections from 73 women who donated breast tissue to the Susan G. Komen for the Cure Tissue Bank at the Indiana University Simon Cancer Center were included in this analysis. Hematoxylin and eosin (H&amp;E) and AE1/AE3 cytokeratin stains were applied to the tissue sections, which were then visually characterized by an expert breast pathologist. H&amp;E stains were used to assess the presence (in ≥5% of the tissue section) or absence of non-proliferative fibrocystic changes, proliferative fibrocystic changes without atypia, inflammatory and reactive conditions, or benign tumors. AE1/AE3 stains were used to quantify the extent of epithelial and stromal content, as well as the number of lobules and ductal structures seen in the section. Quantitative breast percent density (PD%) and whole-breast texture analysis was performed on the digital mammograms using validated software. Parenchymal texture measures recently demonstrated to be associated with breast cancer risk were assessed, including the gray-level co-occurrence features Cluster-shade, Correlation, Energy, Entropy, Inertia and Inverse Difference Moment. Texture was characterized within a 2.5cm2 area of interest in the upper-outer quadrant of the mammograms corresponding to the approximate location of the biopsy. Partial-R2 (pR2) analysis was applied in order to assess the proportion of the variability in the mammographic texture which can be directly attributed to underlying histological properties after adjusting for race, menopausal status, body mass index, PD% and mammography unit manufacturer, utilizing a false-discovery-rate (FDR) multiple comparisons correction. Results: The presence of proliferative fibrocystic changes without atypia was found to have a significant association to Correlation (pR2=0.21), Energy (pR2=0.13), Entropy (pR2=0.16), and Inverse Difference Moment (pR2=0.14) at the FDR-corrected significance level of p&amp;lt;0.0042, even after adjusting for race, menopausal status, body mass index, PD% and the mammography unit manufacturer. These findings suggest that 13-21% of the variation in parenchymal texture patterns seen between women can be explained by the presence of proliferative fibrocystic changes without atypia, a common and well known risk factor for breast cancer. Conclusion: In this study, we identified proliferative fibrocystic changes without atypia as one potential driver of image-based biomarkers of breast cancer risk. Ultimately, identification of radiographic biomarkers of tissue microenvironment could aid in the development of targeted chemopreventative interventions for women with dense breast tissue while simultaneously providing image-based biomarkers to monitor their efficacy. Citation Format: Keller BM, Batiste RC, Chen J, McDonald ES, Conant EF, Kontos D, Feldman MD. Identification of histopathologic determinants of mammographic breast density as a cancer risk factor. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-01-02.

Fast Segmentation of Colour Apple Image under All-Weather Natural Conditions for Vision Recognition of Picking Robots

In order to resolve the poor real-time performance problem of the normalized cut (Ncut) method in apple vision recognition of picking robots, a fast segmentation method of colour apple images based on the adaptive mean-shift and Ncut methods is proposed in this paper. Firstly, the traditional Ncut method based on pixels is changed into the Ncut method based on regions by the adaptive mean-shift initial segmenting. In this way, the number of peaks and edges in the image is dramatically reduced and the computation speed is improved. Secondly, the image is divided into regional maps by extracting the R-B colour feature, which not only reduces the quantity of regions, but also to some extent overcomes the effect on illumination. On this basis, every region map is expressed by a region point, so the undirected graph of the R-B colour grey-level feature is attained. Finally, regarding the undirected graph as the input of Ncut, we construct the weight matrix W by region points and determine the number of clusters based on the decision-theoretic rough set. The adaptive clustering segmentation can be implemented by an Ncut algorithm. Experimental results show that the maximum segmentation error is 3% and the average recognition time is less than 0.7s, which can meet the requirements of a real-time picking robot.

An adaptive two-scale enhancement method to visualize man-made objects in very high resolution SAR images

Ocean vessels and aircrafts usually cause large oscillated responses and span a high dynamic range (HDR) in synthetic aperture radar (SAR) images. To visualize and further recognize these objects on common display devices, it is necessary to compress the HDR at the proper ratio and enhance the details at both low and high grey levels. To achieve this goal, an adaptive two-scale enhancement method is proposed. First, to enhance the details at low grey levels, an adaptive global HDR compression algorithm is designed based on the grey-level distribution feature. Then, with a simplified heterogeneity measurement based on the weighted standard deviation and entropy at a local scale, high-backscattering objects are determined and enhanced to highlight their peak details at high grey levels. The proposed method is validated using two different types of HDR SAR images, and the experimental results suggest that the proposed method visualizes the information at all grey levels and enhances the object’s local peaks. Compared with gamma mapping algorithms, which are widely used for SAR image display, the proposed method outperforms in visualizing the details of man-made objects in very high resolution SAR images.

OFFLINE SIGNATURE VERIFICATION BASED ON GLCM

Several papers have recently appeared in the literature which propose pseudo-dynamic features for automatic static handwritten signature verification based on the use of gray level values from signature stroke pixels. Good results have been obtained using rotation invariant uniform local binary patternsLBP plus LBP and statistical measures from gray levelco-occurrence matrices (GLCM) with MCYT and GPDS offline signature corpuses. In these studies the corpuses contain signatures written on a uniform white “nondistorting” background, however the gray level distribution of signature strokes changes when it is written on a complex background, such as a check or an invoice. The aim of this paper is to measure gray level features robustness when it is distorted by a complex background and also to propose more stable features. A set of different checks and invoices with varying background complexity is blended with the MCYT and GPDS signatures. The blending model is based on multiplication. The signature models are trained with genuine signatures on white background and tested with other genuine and forgeries mixed with different backgrounds. Results show that a basic version of local binary patterns (LBP) or local derivative and directional patterns are more robust than rotation invariant uniform LBP or GLCM features to the gray level distortion when using a support vector machine with histogram oriented kernels as a classifier.

Computerized detection of noncalcified plaques in coronary CT angiography: evaluation of topological soft gradient prescreening method and luminal analysis.

The buildup of noncalcified plaques (NCPs) that are vulnerable to rupture in coronary arteries is a risk for myocardial infarction. Interpretation of coronary CT angiography (cCTA) to search for NCP is a challenging task for radiologists due to the low CT number of NCP, the large number of coronary arteries, and multiple phase CT acquisition. The authors conducted a preliminary study to develop machine learning method for automated detection of NCPs in cCTA. With IRB approval, a data set of 83 ECG-gated contrast enhanced cCTA scans with 120 NCPs was collected retrospectively from patient files. A multiscale coronary artery response and rolling balloon region growing (MSCAR-RBG) method was applied to each cCTA volume to extract the coronary arterial trees. Each extracted vessel was reformatted to a straightened volume composed of cCTA slices perpendicular to the vessel centerline. A topological soft-gradient (TSG) detection method was developed to prescreen for NCP candidates by analyzing the 2D topological features of the radial gradient field surface along the vessel wall. The NCP candidates were then characterized by a luminal analysis that used 3D geometric features to quantify the shape information and gray-level features to evaluate the density of the NCP candidates. With machine learning techniques, useful features were identified and combined into an NCP score to differentiate true NCPs from false positives (FPs). To evaluate the effectiveness of the image analysis methods, the authors performed tenfold cross-validation with the available data set. Receiver operating characteristic (ROC) analysis was used to assess the classification performance of individual features and the NCP score. The overall detection performance was estimated by free response ROC (FROC) analysis. With our TSG prescreening method, a prescreening sensitivity of 92.5% (111/120) was achieved with a total of 1181 FPs (14.2 FPs/scan). On average, six features were selected during the tenfold cross-validation training. The average area under the ROC curve (AUC) value for training was 0.87 ± 0.01 and the AUC value for validation was 0.85 ± 0.01. Using the NCP score, FROC analysis of the validation set showed that the FP rates were reduced to 3.16, 1.90, and 1.39 FPs/scan at sensitivities of 90%, 80%, and 70%, respectively. The topological soft-gradient prescreening method in combination with the luminal analysis for FP reduction was effective for detection of NCPs in cCTA, including NCPs causing positive or negative vessel remodeling. The accuracy of vessel segmentation, tracking, and centerline identification has a strong impact on NCP detection. Studies are underway to further improve these techniques and reduce the FPs of the CADe system.

A Hybrid Approach to Diagnosis of Hepatic Tumors in Computed Tomography Images

Liver cancer is one of the most popular cancer diseases and causes a large amount of death every year, can be reduced by early detection and diagnosis. Computer-aided liver analysis can help in the early detection and diagnosis of liver cancer. In this paper, enhancement and segmentation process is applied to increase the computation and focus on liver parenchyma. This parenchyma also segmented using Watershed and Region Growing algorithms to extract liver tumors. These tumors will be analyzed and characterized to distinguish between hemangioma (benign) and hepatocellular (malignant) tumors using Local Binary Pattern (LBP), Gray Level Co-occurrence matrix (GLCM), Fractal Dimension (FD) and feature fusion technique is applied to maximize and enhance the performance of the classifier rate. The authors review different methods for liver segmentation and abnormality classification. An attempt was made to combine the individual scores from different techniques in order to compensate their individual weaknesses and to preserve their strength. The authors present and exhaustively evaluate algorithms using computer vision techniques. The experimental results based on confusion matrix and kappa coefficient show that the higher accuracy is obtained of automatic agreement classification and suggest that the developed CAD system has great potential and promise in the automatic diagnosis of both benign and malignant tumors of liver.

License Plate Recognition Model Research Based on the Multi-Feature Technology

Due to the impact of pollution, environment and so on in actual scene, it is difficult for the traditional single feature recognition model to obtain a higher accuracy of the license plate recognition. The paper proposed a new license plate image recognition model. First, the structural features and gray level features of the license plate, such as the contour and stroke order are extracted. Then, the principal component analysis is used to carry out the fusion, dimensionality reduction and redundancy removal processing for the two kinds of features, and a fuzzy fusion model for differentiated features is introduced to ensure minimal loss for the feature in fusion. Finally, the final result of the license plate image recognition is achieved in accordance with high degree of confidence criterion when the slope interference of the license plate is considered fully. Simulation results show that the license plate image recognition model based on multi-feature combination can solve the problem of the single feature recognition model, improve the accuracy of license plate recognition which is up to 99%. Moreover, the model has a faster recognition speed and can be applied to the actual license plate recognition. DOI : http://dx.doi.org/10.11591/telkomnika.v12i4.4307

PERFORMANCE ANALYSIS OF BRAIN TUMOR DIAGNOSIS BASED ON SOFT COMPUTING TECHNIQUES

Computed tomography images are widely used in the diagnosis of brain tumor because of its faster processing, avoiding malfunctions and suitability with physician and radiologist. This study proposes a new approach to automated detection of brain tumor. This proposed work consists of various stages in their diagnosis processing such as preprocessing, anisotropic diffusion, feature extraction and classification. The local binary patterns and gray level co-occurrence features, gray level and wavelet features are extracted and these features are trained and classified using Support vector machine classifier. The achieved results and quantitatively evaluated and compared with various ground truth images. The proposed method gives fast and better segmentation and classification rate by yielding 99.4% of sensitivity, 99.6% of specificity, 97.03% of positive predictive value and 99.5% of overall accuracy.

Abstract P4-03-04: Computer extracted image measurements of nuclear shape and texture from H&amp;E images appear to stratify low and high risk ER+ breast cancers assessed via oncotype DX

Abstract Background: In this study we investigate the ability of computer extracted image features (nuclear morphology and texture) from digitized H&amp;E tissue slides to stratify women with lymph node negative (LN-), estrogen receptor positive (ER+) breast cancer (BCa) as low or high risk as determined by Oncotype DX (ODX), a 21 gene-expression assay. Each year, over 120,000 women in the United States (1 million worldwide) are diagnosed with ER+ BCa. Treatment guidelines recommend hormone therapy (HT) plus chemotherapy (CT); however, up to 85% of ER+ BCa patients will not benefit from CT, yet will still suffer its side effects. ODX yields a numeric risk score (RS) ranging from 1-100; RS&amp;lt;18 suggests patients will respond to HT alone while RS&amp;gt;30 indicates need for adjuvant CT. Unfortunately, this test is expensive (&amp;gt;$4000), time-consuming, and involves destructive tissue testing. The goal of this study is to show that quantitative features calculated from H&amp;E images can accurately predict risk stratification as determined by ODX in women with LN-, ER+ BCa, suggesting a histologic image based classifier could serve as a low-cost alternative. Methods: Digitized H&amp;E-stained ER+ BCa tissue sampled from 111 patients (34 high and 77 low-risk as determined by ODX) were obtained from the University of Pennsylvania, the University of Medicine and Dentistry of NJ, and Case Western Reserve University. Regions of cancer were annotated manually by an expert pathologist, and representative fields of view (FOV) were chosen at 20x magnification (2000 by 2000 pixels) for each patient. A selection of nuclear boundaries was annotated manually in each FOV. For each nucleus, a set of 2343 features was extracted, including 21 morphological (size, shape, and boundary) and 2322 texture (Gabor, Local Binary Pattern, Greylevel, and Laws filter features). Using Minimum Redundancy Maximum Relevance (mRMR) feature selection, the 3 features best able to separate low and high ODX risk categories were identified and used to build a supervised Bayesian classifier. Classifier training employed a randomized 3-fold cross-validation scheme; in each trial, two-thirds of the dataset were randomly selected for training, and the remaining one-third employed for independent testing. Classifier performance was evaluated using area under the receiver operating characteristic curve (AUC), positive predictive value (PPV), and negative predictive value (NPV) with respect to low and high ODX risk categorization. Performance metrics were averaged over 100 trials of 3-fold cross-validation (see table). Results: The mRMR method selected one morphological feature (nuclear area) and two Laws-based texture features as being highly discriminating between risk categories. The Bayesian classifier trained with these 3 features yielded high AUC, PPV, and NPV measures with low variance in distinguishing ODX risk categories. The supervised classification results indicate that quantitative image features from H&amp;E-stained histopathology are able to accurately discriminate between low and high risk patients as determined by ODX. Classification PerformancePerformance MetricAverage (100 Trials)Standard DeviationAUC0.870.018PPV0.810.039NPV0.880.017 Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-03-04.

Research on Image Technology with Algorithm of Image Threshold Segmentation Based on Gray Level Characteristics

In the process of images analyzing, some research areas are usually catches people’s eyes which has unique properties called objects or prospects; therefore the research on how to get these objects becomes the center of attention. Segmentation method based on gray threshold can quickly distinguish foreground from the background in the image and provide a good foundation for image detection and process. Two kinds of threshold segmentation method based on the gray level features in the image processing and its defects are mainly discussed in this paper, and the algorithm to get optimal threshold to improve the deficiency of the former methods was proposed in application.

Underwater Image Transition Region Extraction and Segmentation Based on SVM

By analyzing the gray level features of transition region, a new underwater image transition region extraction method based on Support Vector Machine (SVM) is presented. At first, a vector is constructed to fully describe the transition region, which includes local complexity, local difference and neighborhood homogeneity. Then, SVM is applied to train and classify the set of feature vectors, so that the transition region of the underwater image is extracted. Finally, the segmentation threshold is determined by mean of the histogram of the transition region, and the binary result was yielded. The experimental results show that the proposed algorithm can achieve a better transition region extraction and segmentation performance, and automatically select the optimal threshold for transition region extraction.