COMPUTER-AIDED BREAST-CANCER DETECTION AND DIAGNOSIS OF MASSES USING DIFFERENCE OF GAUSSIANS AND DERIVATIVE-BASED FEATURE SALIENCY

A new model-based vision (MBV) algorithm is developed to find regions of interest (ROI's) corresponding to masses in digitized mammograms an d to classify the masses as malignant/benign, The MBV algorithm is com prised of five modules to structurally identify suspicious ROI's, elim inate false positives, and classify the remaining as malignant or beni gn. The focus of attention module uses a difference of Gaussians (DoG) filter to highlight suspicious regions in the mammogram, The index mo dule uses tests to reduce the number of nonmalignant regions from 8.39 to 2.36 per full breast image. Size, shape, contrast, and Laws textur e features are used to develop the prediction module's mass models, De rivative-based feature saliency techniques are used to determine the b est features for classification, Nine features are chosen to define th e malignant/benign models, The feature extraction module obtains these features from all suspicious ROI's, The matching module classifies th e regions using a multilayer perceptron neural network architecture to obtain an overall classification accuracy of 100% for the segmented m alignant masses with a false-positive rate of 1.8 per full breast imag e. This system has a sensitivity of 92% for locating malignant ROI's, The database contains 272 images (12 b, 100 mu m) with 36 malignant an d 53 benign mass images. The results demonstrate that the MBV approach provides a structured order of integrating complex stages into a syst em for radiologists.