Automatic Breast Tumor detection and classification of Asymmetries in Mammograms using Neural Network Classifier and Hybrid GA

Breast Cancer has become a common mortality factor in India. Despite the fact, not all general hospitals have the mammogram facilities. The long waiting for diagnosing a breast cancer may increase the possibility of fatality and the cancer spreading. Therefore a computerized breast cancer may increase diagnosis prototype has been developed to reduce the time taken and indirectly reducing the probability of death. Micro calcification on X-ray mammogram is a significant mark for early detection of breast cancer. In this paper, A hybrid Genetic algorithm and Neural Network is proposed to automatically detect the suspicious regions on digital mammograms based on irregularity between left and right breast image. The basic idea of Irregularity approach is corresponding left and right images are subtracted to extract the suspicious region. The proposed system consists two steps: First, the mammogram images are enhanced using the median filter, pectoral muscle region is removed and the border of the mammogram is detected for the both left and right images of the binary image.