Breast Mass Classification using Statistical and Local Binary Pattern Features

Millions of women are suffering from breast cancer today. Breast cancer can be treated effectively when detected early. Mammography is broadly recognized as the most effective imaging modality for an early detection of breast cancer abnormalities. Computer-aided diagnosis systems are very helpful for radiologist in detection and diagnosing abnormalities earlier and faster than traditional screening programs. In this paper, two techniques are proposed based on statistical and LBP features using support vector machine (SVM) and the k-nearest neighbor (KNN) classifiers. The evaluation of the system is applied on Digital Database for Screening Mammography (DDSM). The system classifies normal from abnormal cases with high accuracy rate.