Implementation of practical computer aided diagnosis system for classification of masses in digital mammograms

Breast cancer is the most common cancer in women worldwide. It is also the principle cause of death from cancer among women globally. Mammogram image is considered as the most reliable, low cost, and highly sensitive technique for detecting small lesions. Computer-aided diagnosis system (CAD) can be very helpful for radiologist in detection and diagnosing abnormalities earlier and faster than traditional screening programs. In this work a CAD system to distinguish between masses and normal breast tissue was proposed. We started our system by using DDSM database for mammogram images which were first preprocessed using image enhancement algorithm, then 100 regions of interest (ROIs) containing masses and normal breast tissue are extracted. Then we extracted a group of 59 texture and statistical features from the ROIs. Then we performed feature selection using Sequential Forward Selection and Sequential Floating Forward Selection. Finally we used K-Nearest Neighbor (KNN) classifier, Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), and Support Vector Machine (SVM) classifier for classification with leave-one-out method for testing. The obtained results show acceptable sensitivity and specificity for the system.