Characterization of Architectural Distortion in Mammograms Based on Texture Analysis Using Support Vector Machine Classifier with Clinical Evaluation.

Chapter 6 - Optimization of ROI Size for Development of Computer Assisted Framework for Breast Tissue Pattern Characterization Using Digitized Screen Film Mammograms

Architectural distortion is an important sign of early breast cancer. We present methods for computer-aided detection of architectural distortion in mammograms acquired prior to the diagnosis of breast cancer in the interval between scheduled screening sessions. Potential sites of architectural distortion were detected using node maps obtained through the application of a bank of Gabor filters and linear phase portrait modeling. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases, including 301 true-positive ROIs, and from 52 mammograms of 13 normal cases. Each ROI was represented by three types of entropy measures of angular histograms composed with the Gabor magnitude response, angle, coherence, orientation strength, and the angular spread of power in the Fourier spectrum, including Shannon's entropy, Tsallis entropy for nonextensive systems, and Rényi entropy for extensive systems. Using the entropy measures with stepwise logistic regression and the leave-one-patient-out method for feature selection and cross-validation, an artificial neural network resulted in an area under the receiver operating characteristic curve of 0.75. Free-response receiver operating characteristics indicated a sensitivity of 0.80at 5.2 false positives (FPs) per patient. The proposed methods can detect architectural distortion in prior mammograms taken 15months (on the average) before clinical diagnosis of breast cancer, with a high sensitivity and a moderate number of FPs per patient. The results are promising and may be improved with additional features to characterize subtle abnormalities and larger databases including prior mammograms.

This paper presents methods for the detection of architectural distortion in mammograms of interval-cancer cases taken prior to the diagnosis of breast cancer, using Gabor filters, phase portrait analysis, fractal dimension (FD), and analysis of the angular spread of power in the Fourier spectrum. In the estimation of FD using the Fourier power spectrum, only the distribution of power over radial frequency is considered; the information regarding the angular spread of power is ignored. In this study, the angular spread of power in the Fourier spectrum is used to generate features for the detection of spiculated patterns related to architectural distortion. Using Gabor filters and phase portrait analysis, a total of 4224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases, including 301 ROIs related to architectural distortion, and from 52 mammograms of 13 normal cases. For each ROI, the FD and measures of the angular spread of power were computed. Feature selection was performed using stepwise logistic regression. The best result achieved, in terms of the area under the receiver operating characteristic curve, is 0.75 ± 0.02 with an artificial neural network including radial basis functions. Analysis of the performance of the methods with free-response receiver operating characteristics indicated a sensitivity of 0.82 at 7.7 false positives per image.

This paper presents methods for the detection of architectural distortion in mammograms of interval-cancer cases taken prior to the diagnosis of breast cancer, using Rényi entropy. Initial candidates for sites of architectural distortion were detected using a bank of Gabor filters and phase portrait analysis. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases, including 301 true-positive ROIs, and from 52 mammograms of 13 normal cases. Each ROI was represented by the Rényi entropy of angular histograms composed with the Gabor magnitude response, angle, coherence, orientation strength, and the angular spread of power in the Fourier spectrum. Using the stepwise logistic regression and leave-one-image-out methods for feature selection, the best results achieved, in terms of the area under the receiver operating characteristic curve, are 0.72 with Fisher linear discriminant analysis and the Bayesian classifier, and 0.75 with an artificial neural network based on radial basis functions. Analysis of the free-response receiver operating characteristics indicated a sensitivity of 0.80 at 7.1 false positives per image.

Optimizing Region of Interest Size and Placement for Opportunistic CT

The analysis of mammograms is a difficult task due to the subtle appearance of some lesions. Computer-aided diagnosis (CAD) have been shown to improve the sensitivity of detection of masses and calcifications; however, there is a need for dedicated methods to detect architectural distortion in the absence of a central mass. Improvement in the detection of architectural distortion may be expected to result in better prognosis for patients with early stages of breast cancer. By employing the concept of phase portraits, a method to characterize architectural distortion in mammograms using texture orientation fields is presented. The results obtained show that the proposed technique can achieve good discrimination between architectural distortion and other parenchymal patterns. Such a technique would be applicable, in a CAD system, to images that have cleared the stages of detection of calcifications and masses with no positive findings.

Metabolic brain imaging with 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET) is a supportive diagnostic and differential diagnostic tool for neurodegenerative dementias. In the clinic, scans are usually visually interpreted. However, computer-aided approaches can improve diagnostic accuracy. We aimed to build two machine learning classifiers, based on two sets of FDG PET-derived features, for differential diagnosis of common dementia syndromes. We analyzed FDG PET scans from three dementia cohorts [63 dementia due to Alzheimer's disease (AD), 79 dementia with Lewy bodies (DLB) and 23 frontotemporal dementia (FTD)], and 41 normal controls (NCs). Patients' clinical diagnosis at follow-up (25 ± 20 months after scanning) or cerebrospinal fluid biomarkers for Alzheimer's disease was considered a gold standard. FDG PET scans were first visually evaluated. Scans were pre-processed, and two sets of features extracted: (1) the expressions of previously identified metabolic brain patterns, and (2) the mean uptake value in 95 regions of interest (ROIs). Two multi-class support vector machine (SVM) classifiers were tested and their diagnostic performance assessed and compared to visual reading. Class-specific regional feature importance was assessed with Shapley Additive Explanations. Pattern- and ROI-based classifier achieved higher overall accuracy than expert readers (78% and 80% respectively, vs. 71%). Both SVM classifiers performed similarly to one another and to expert readers in AD (F1 = 0.74, 0.78, and 0.78) and DLB (F1 = 0.81, 0.81, and 0.78). SVM classifiers outperformed expert readers in FTD (F1 = 0.87, 0.83, and 0.63), but not in NC (F1 = 0.71, 0.75, and 0.92). Visualization of the SVM model showed bilateral temporal cortices and cerebellum to be the most important features for AD; occipital cortices, hippocampi and parahippocampi, amygdala, and middle temporal lobes for DLB; bilateral frontal cortices, middle and anterior cingulum for FTD; and bilateral angular gyri, pons, and vermis for NC. Multi-class SVM classifiers based on the expression of characteristic metabolic brain patterns or ROI glucose uptake, performed better than experts in the differential diagnosis of common dementias using FDG PET scans. Experts performed better in the recognition of normal scans and a combined approach may yield optimal results in the clinical setting.

The problem with development of the support vector machine (SVM) classifiers using modified particle swarm optimization (PSO) algorithm and their ensembles has been considered. Solving this problem would allow fulfilling the high-precision data classification, especially Big Data classification, with the acceptable time expenditures. The modified PSO algorithm conducts a simultaneous search of the type of kernel functions, the parameters of the kernel function and the value of the regularization parameter for the SVM classifier. The idea of particles' «regeneration» served as the basis for the modified PSO algorithm. In the implementation of this algorithm, some particles change the type of their kernel function to the one which corresponds to the particle with the best value of the classification accuracy. The offered PSO algorithm allows reducing the time expenditures for the developed SVM classifiers, which is very important for Big Data classification problem. In most cases such SVM classifier provides the high quality of data classification. In exceptional cases the SVM ensembles based on the decorrelation maximization algorithm for the different strategies of the decision-making on the data classification and the majority vote rule can be used. Also, the two-level SVM classifier has been offered. This classifier works as the group of the SVM classifiers at the first level and as the SVM classifier on the base of the modified PSO algorithm at the second level. The results of experimental studies confirm the efficiency of the offered approaches for Big Data classification.

In breast X-ray imaging, breast texture has been characterized by a radial noise power spectrum (NPS) that has an inverse power-law shape with exponent β. The technique to estimate the radial power-law coefficient β is typically based on averaging 2-dimensional noise power spectra (NPS), calculated from partly overlapping image regions each weighted by a suitable window function. The linear regression applied over a selected frequency range to the logarithm of the 1- dimensional NPS as a function of the logarithm of the radial frequencies, gives β. For each step in this process, several alternative techniques have been proposed. This paper investigates the effect of image region of interest (ROI) size, image data windowing and alternative ways to determine radial frequency in terms of bias, variance and root mean square error (RMSE) in the estimated β. The effects of these three factors were analytically derived and evaluated using synthetic images with known β varying from 1 to 4 to cover the range of textures encountered in 2D and 3D breast X-ray imaging. Our results indicate that the RMSE in estimated β is smallest when the ROIs are multiplied with an appropriate window function and either no radial averaging or radial averaging with small frequency bins is applied. The ROI size yielding the smallest RMSE depends on several factors and needs to be validated with numerical simulations. In clinical practice however, there might be a need to compromise in the choice of the ROI size to balance between the RMSE magnitudes inherent to the applied β estimation technique and encompass the breast texture range so as to obtain an accurate shape of the NPS. When using 2.56 cm x 2.56 cm ROI sizes, applying a 2D Hann window and no radial frequency averaging, the RMSE in the estimated β ranges from 0.04 to 0.1 for true β values equal to 1 and 4. While many subtleties in real images were not modeled to simplify the mathematics in deriving our results, this work is illustrative in demonstrating the limits of commonly used algorithm steps to estimate accurate β values.

The accurate preoperative histopathological grade diagnosis of adult gliomas is of great significance for the formulation of a surgical plan and the implementation of a subsequent treatment. The aim of this study is to establish a predictive model for classifying adult gliomas into grades 2-4 based on preoperative conventional multimodal MRI radiomics. Patients with pathologically confirmed gliomas at Huashan Hospital, Fudan University, between February 2017 and July 2019 were retrospectively analyzed. Two regions of interest (ROIs), called the maximum anomaly region (ROI1) and the tumor region (ROI2), were delineated on the patients' preoperative MRIs utilizing the tool ITK-SNAP, and Pyradiomics 3.0 was applied to execute feature extraction. Feature selection was performed utilizing a least absolute shrinkage and selection operator (LASSO) filter. Six classifiers, including Gaussian naive Bayes (GNB), random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM) with a linear kernel, adaptive boosting (AB), and multilayer perceptron (MLP) were used to establish predictive models, and the predictive performance of the six classifiers was evaluated through five-fold cross-validation. The performance of the predictive models was evaluated using the AUC and other metrics. After that, the model with the best predictive performance was tested using the external data from The Cancer Imaging Archive (TCIA). According to the inclusion and exclusion criteria, 240 patients with gliomas were identified for inclusion in the study, including 106 grade 2, 68 grade 3, and 66 grade 4 gliomas. A total of 150 features was selected, and the MLP classifier had the best predictive performance among the six classifiers based on T2-FLAIR (mean AUC of 0.80 ± 0.07). The SVM classifier had the best predictive performance among the six classifiers based on DWI (mean AUC of 0.84 ± 0.05); the SVM classifier had the best predictive performance among the six classifiers based on CE-T1WI (mean AUC of 0.85 ± 0.06). Among the six classifiers, based on ROI1, the MLP classifier had the best prediction performance (mean AUC of 0.78 ± 0.07); among the six classifiers, based on ROI2, the SVM classifier had the best prediction performance (mean AUC of 0.82 ± 0.07). Among the six classifiers, based on the multimodal MRI of all the ROIs, the SVM classifier had the best prediction performance (average AUC of 0.85 ± 0.04). The SVM classifier, based on the multimodal MRI of all the ROIs, achieved an AUC of 0.81 using the external data from TCIA. The prediction model, based on preoperative conventional multimodal MRI radiomics, established in this study can conveniently, accurately, and noninvasively classify adult gliomas into grades 2-4, providing certain assistance for the precise diagnosis and treatment of patients and optimizing their clinical management.

A case study on computer-aided diagnosis of nonerosive reflux disease using deep learning techniques

SUMMARYNeuronal oscillations are suggested to play an important role in auditory working memory (WM), but their contribution to content-specific representations has remained unclear. Here, we measure magnetoencephalography during a retro-cueing task with parametric ripple-sound stimuli, which are spectrotemporally similar to speech but resist non-auditory memory strategies. Using machine learning analyses, with rigorous between-subject cross-validation and non-parametric permutation testing, we show that memorized sound content is strongly represented in phase-synchronization patterns between subregions of auditory and frontoparietal cortices. These phase-synchronization patterns predict the memorized sound content steadily across the studied maintenance period. In addition to connectivity-based representations, there are indices of more local, “activity silent” representations in auditory cortices, where the decoding accuracy of WM content significantly increases after task-irrelevant “impulse stimuli.” Our results demonstrate that synchronization patterns across auditory sensory and association areas orchestrate neuronal coding of auditory WM content. This connectivity-based coding scheme could also extend beyond the auditory domain.

Glioma grading between intermediate grades (Grade II vs. III and Grade III vs. IV) as well as multiclass grades (Grade II vs. III vs. IV) is challenging and needs to be addressed. To develop an artificial intelligence-based methodology for glioma grading using T1 perfusion parameters and volume of tumor components, and validate the efficacy of the methodology by grading on a cohort of glioma patients. Retrospective. The development set consisted of 53 glioma patients and validation consisted of 13 glioma patients. Conventional MRI images (2D T1 -W, dual PD-T2 -W, and 3D FLAIR) and 3D T1 perfusion MRI data obtained at 3 T. Enhancing and nonenhancing components of glioma were segmented out and combined to form the region of interest (ROI) for glioma grading. Prominent vessels were removed from the selected ROI. Different T1 perfusion parameters from the ROI were combined with volume of tumor components to form the feature set for glioma grading. Optimization was carried out for selection of the statistic of the T1 perfusion parameters and the features to be used for glioma grading using sequential feature selection and random forest-based feature selection method. An optimized support vector machine (SVM) classifier was used for glioma grading. Mean ± SD, analysis of variance (ANOVA) followed by the Tukey-Kramer test, ROC analysis. Classification error for Grade II vs. III was 3.7%, for Grade III vs. IV was 5.26%, and for Grade II vs. III vs. IV was 9.43% using the proposed methodology. The mean of the values above the 90th percentile value of T1 perfusion parameters provided a maximum area under the curve (AUC) for intermediate grade differentiation. Random forest obtained optimal feature set provided better grading results than other methods using the SVM classifier. It was feasible to achieve low classification error for intermediate as well as multiclass glioma grading using an SVM classifier based on optimized features obtained from T1 perfusion MRI and volumes of tumor components. 4 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2019;50:1295-1306.

A boosted SVM classifier trained by incremental learning and decremental unlearning approach

In this paper, a liver diseases diagnosis based on Gabor wavelets and support vector machine (SVM) classifier is proposed. The diagnosis scheme includes two steps: features extraction and classification. The features derived from Gabor wavelets are obtained from the regions of interest (ROIs) among the normal and abnormal CT images. In the classification step, the SVM classifier is used to discriminate the fiver diseases. Finally the receiver operating characteristic (ROC) curve is employed to evaluate the performance of the diagnosis system. Three kinds of fiver diseases are identified including cyst, hepatoma and cavernous hemangioma. The effectiveness of the proposed method is demonstrated through experimental results on CT images including 76 fiver cysts, 30 hepatomas, and 40 cavernous hemangiomas. From the results we can observe that the discrimination rate of cyst is higher than other diseases, and the classification accuracy decreases slightly between cavernous hemangiomas and hepatomas. However, a normal region can be discriminated from all of these diseases at most entirety.