Detection of Breast Cancer in Mammogram using Support Vector Machine

Varsha J. Gaikwad

Abstract: Breast cancer is leading cause of death in women. An early sign of breast cancer is the presence of a tiny deposit of calcium i.e microcalcifications in mammogram. To detect this micro-calcification the Support Vector Machine (SVM) based classifier techniques is used. In this paper we proposed the technique to detect the cancer in the mammogram. The proposed method has been implemented in four stages preprocessing, segmentation for ROI extraction, feature extraction and classification. The proposed method was evaluated with Mammogram Image Analysis Society (MIAS) database. Experimental results show that, SVM shows better accuracy.

Keywords: Breast Cancer, Mammography, preprocessing, marker-controlled segmentation, SVM classifier