Breast Cancer Detection using Two Dimensional Principal Component Analysis and Back Propagation Neural Network

Abstract

Breast cancer is the most common cancer which affects women around the world. It has been increasing over the years. Detection and diagnosis is the main factor for breast cancer control which increases the success rate of treatment, saves lives and reduce the cost. This paper proposes an efficient approach for breast cancer detection in mammogram breast images using two dimensional principal component analysis and back propagation neural network. The proposed approach consists of four step by step procedures namely preprocessing of breast images, image enhancement, feature extraction and classification. Two dimensional principal component analysis is used to obtain the features of the preprocessed and enhanced image. The reason for selecting two dimensional principal component analysis is it is easier to evaluate the covariance matrix accurately and less time is required to determine the corresponding features. Finally, Back propagation neural network is used to classify whether the given mammogram image is normal or abnormal. Simulation results are carried out using the proposed approach by considering MIAS data base. From the results, it is observed that proposed approach provide better accuracy.