Computer aided diagnosis system for abdomen diseases in computed tomography images

摘要: Abstract In this paper, a computer aided diagnostic (CAD) system for classification of abdomen diseases from computed tomography (CT) images is presented. The methodology used in paper to select the most appropriate machine learning technique segmentation, feature extraction and each module proposed CAD. selecting CAD results accurate efficient system. Regions interest are segmented CT tumor, cyst, calculi normal liver using active contour models, region growing thresholding. presented research work exploits discriminating power features classifying abdominal diseases. Therefore, extracts statistical texture descriptors three kinds methods i.e. Gray-Level co-occurrence matrices (GLCM), Discrete Wavelet Transform (DWT) Curvelet (DCT). At next stage, effective optimum ROIs selected Genetic Algorithm (GA). Further, Support Vector Machine (SVM) Artificial Neural Network (ANN) assess capability abdomen. study performed on 120 (30 normal, 30 cyst calculi). It observed that consists edge based model combined with optimized DCT along ANN as classifier achieves best performance 95.1%. also shown highest sensitivity, specificity 95% 98% respectively.