Non Invasive Detection of Breast Cancer and Feature Selection from Mammogram Images

Abstract

Mammography is the main test used for screening the breast cancer. It use low energy x-ray for screening. Hence mammography is typically recommended for early diagnosis and provides a key to improve breast cancer detection and estimating the prognosis. Once the symptoms of breast cancer are detected, doctor suggests for biopsy. Based on Histology and mammogram image cancer will be graded. The objective of this paper is detection of breast cancer using Mammogram image itself without using biopsy. The procedure consists of pre-processing, segmentation and feature extraction. Three types of segmentation methods were handled and the effective method is estimated. Shape and textural features were extracted from cancer image and correlated clinically.