Multiple TBSVM-RFE for the detection of architectural distortion in mammographic images

摘要

Breast cancer is a leading health threaten for women in the world. Among the several abnormalities observable on mammograms, architecture distortion is one of the most difficult to detect due to its subtlety. Computer-Aided Diagnosis (CAD) technology has been widely used for the detection and diagnosis of breast cancer. In this paper, a new automatic architectural distortion detection method for breast cancer in mammographic images is proposed. Firstly, Gabor filters and phase portrait analysis are used to locate the suspicious regions based on the image characteristic of architectural distortion. Twin bounded Support Vector Machine (TBSVM) is employed to reduce the large amounts of false positives. TBSVM is a kind of binary classifier, which has advantages in both computation efficiency and generalization when dealing with binary classification. For each suspicious region, several features are extracted. However, not every extracted feature contributes to the classification accuracy. We proposed a novel feature selection method for TBSVM and utilized it for the architectural distortion detection in mammograms, named Multiple Twin Bound Support Vector Machines Recursive Feature Elimination (MTBSVM-RFE). The results showed that our proposed method detect the region of architecture distortion with high accuracy.