Microcalcifications are one of the earliest signs of breast cancer. Physicians generally agree that mammography is the best technique for detecting microcalcification. Microcalcification is too small to detect by palpable breast diagnosis.
In this dissertation, we propose an adaptive microcalcification detection method in mammography. In the proposed method, an input mammogram is preprocessed by mammogram background detection, boundary detection and image enhancement method with homomorphic filtering in wavelet domain. Adaptive denoising using wavelet shrinkage is also performed in this part. To improve the visibility of breast cancer, various image enhancement methods have been performed in previous studies. The aim of these studies was to increase the contrast of microcalcification. The proposed adaptive denoising algorithm takes the noise characteristics of each mammogram into the process. For a mammogram, its noise characteristics are obtained by analyzing the background region. The adaptive enhancement and denoising method shows better visibility of mammogram than the previous methods. Noise effect is reduced significantly while microcalcifications are more clearly seen.
The enhanced mammogram is then processed in the next part of CAD system to detect microcalcification. The detection system has two stages where the first stage finds potential microcalcification pixels (ROI) and the second one detects the microcalcification within the ROIs. Both stages use ANN and SVM for the purpose of detection. The detected pixels from stage 1 are clustered to regions of interest (ROI). And the second stage decides which ROIs are malignant microcalcifications.
Experiment is performed with three kinds of image enhancement methods: the proposed adaptive enhancement method, previous homomorphic-filtering method, and histogram-stretching method. Experimental results are illustrated by the free-response operation characteristics (FROC) curves. FROC curves indicate...