ROI Detection in Mammogram Images Using Wavelet-Based Haralick and HOG Features

Abstract

Digital mammography is a widespread medical imaging tech-nique that is used for early detection and diagnosis of breast cancer. Detecting the region of interest (ROI) helps to locate the abnormal areas, which may be analyzed further by a ra-diologist or a CAD system. In this paper, a new classification method is proposed for ROI detection in mammography im-ages. Features are extracted using Wavelet transform, Haralick and HOG descriptors. To reduce the number of di-mensions and eliminate irrelevant features, a wrapper-based feature selection method is implemented. Several feature ex-traction methods and machine learning classifiers are com-pared by performing a leave-one-image-out cross-validation experiment on a difficult dataset. The proposed feature ex-traction method provides the best accuracy of 87.5% and the second-best area under curve (AUC) score of 84% when em-ployed in a random forest classifier.