Early Stage Breast Cancer Detection through Mammographic Feature Analysis

Breast cancer is the second leading cause of cancer amongst women. Mammography plays a very important role in early stage detection of breast cancer. Computer aided design (CAD) systems are used to assist radiologists in better classification of tumor in a mammograph as benign or malignant. For early stage detection of breast cancer CAD systems require features extracted from mammographs. A new feature-set was formed involving six preexisting and one devised feature. Thirty-three images from Mini-mias database were selected for this study. The cases included 16 circumscribed benign, 4 circumscribed malignant, 9 speculated benign, and 5 speculated malignant lesions. The features were trained using Kohnan neural networks. Results show 80% classification rate.