Implementation of Pectoral Removal with Hybridization of MSER Pectoral and Hough Pectoral Methods

In women worldwide, breast cancer is the most frequent disease and the second greatest death cause. Many lives are being lost as a result of breast cancer. Mammography is the most widely used screening method for breast cancer. It gives mammogram image of breast organ. MLO view is one kind of mammogram view. We can see masses in MLO mammogram which is main indicator of the breast cancer. Computer-aided systems are being used in the detection of mass in the MLO mammogram. The MLO mammogram includes a portion of the pectoral muscle also. Developing a computer-aided system to detect the mass is a challenging task, and the presence of pectoral tissue in a mammogram has a detrimental effect on the identification of masses because its grey levels are similar to those of masses. Therefore, pectoral removal considerably enhances mass detection. This paper analyses the performance of the hybrid computer-aided diagnosis technique “ensemble of Hough pectoral and MSER pectoral methods” for removing the Pectoral area. The results demonstrate that the suggested approach improves the precision of pectoral region removal. The proposed technique has an accuracy of 98.1% compared to the average 95% accuracy of existing methods. The mini-MIAS database is used for performance evaluation.