A Method of Detection Micro-Calcifications in Mammograms Using Wavelets and Adaptive Thresholds

Breast cancer is one of the most common malignant diseases among women, it is important to give patients early diagnose and treatment. Mammography has become the most effective way for detection of breast cancer, and it is sensitive to clustered micro-calcification which is the key characteristic of early breast tumors. In this paper, we propose a method of detection micro-calcification. We first select the regions of interest (ROI) from the whole breast area by using wavelet and adaptive thresholds according to each mammogram, which are the doubtful micro-calcification regions; then the ROIs are further analyzed by DOG filter to reduce false positive rate. Experimental results indicate that the proposed method can provide good detection performance.