Segmentation of mammogram abnormalities using ant system based contour clustering algorithm

Breast cancer is the most widespread cancer that affects females all over the world. The Computer-aided Detection Systems (CADs) could assist radiologists’ in locating and classifying the breast tissues into normal and abnormal, however the absolute decisions are still made by the radiologist. In general, CAD system consists of four stages: Pre-processing, segmentation, feature extraction, and classification. This research work focuses on the segmentation step, where the abnormal tissues are segmented from the normal tissues. There are numerous approaches presented in the literature for mammogram segmentation. The major limitation of these methods is that they have to test each and every pixel of the image at least once, which is computationally expensive. This research work focuses on detection of microcalcifications from the digital mammograms using a novel segmentation approach based on novel Ant Clustering approach called Ant System based Contour Clustering (ASCC) that simulates the ants’ foraging behavior. The performance of the ASCC based segmentation algorithm is investigated with the mammogram images received from Mammographic Image Analysis Society (MIAS) database.