A Computer Aided Detection System for Breast Cancer in the MammogramsBased on Particle Swarm Optimization Algorithm

Abstract

the majority cancer mortality among women is due to breast cancer over the world wide. Recent researches have shown the effectiveness of x-ray mammography in early detection of breast cancer. Unfortunately, the present systems for early detection are expensive and needs extremely complex algorithms. The crucial challenge in designing a computer-aided detection (CAD) systems for breast cancer are the segmentation phase, which requires highly complex computation. Hence, this paper proposes a CAD system to be utilized for breast cancer detection in mammographic datasets. The segmentation step is performed by a Particle Swarm Optimization Algorithm (PSO). Statistical, textural and shape feature are calculated over the segmented region. A non linear support vector machine (SVM) is exploited in the next phase in order to analyze the extracted features and classify the mammograms into normal, benign or malignant. For the sack of evaluating the performance, the experiment is performed on Mini-MIAS database. The obtained accuracy rates based on 10-folds cross validation are 85.4% for classifying normal from abnormal, 89.5% for classifying malignant from benign. The experiment shows that the classification accuracy is 81% when classifying normal, malignant or benign. The result compromises with recent researches concurs that the proposed algorithm compromises between the achieved accuracy to complexity cost.