Fuzzy logic and scale space approach to microcalcification detection

Breast cancer is one of the leading causes of female mortality. Since the causes are unknown, breast cancer cannot be prevented. Microcalcifications and masses are the earliest signs of breast carcinomas and their detection is one of the key issues for breast cancer control. It is difficult for radiologists to provide accurate and uniform evaluations of the enormous number of mammograms generated in widespread screening. Computer-aided detection of microcalcifications and masses is an important and challenging task. The paper presents a novel approach for detecting microcalcification clusters (MCCs). First, mammograms are normalized. Then, fuzzy set theory and the fuzzy entropy principle are used to fuzzify the mammograms and the fuzzified images are enhanced. Finally, scale-space and Laplacian-of-a-Gaussian filter techniques are used to determine the sizes and locations of microcalcifications. A free-response operating characteristic (FROC) curve is used to evaluate the performance. The major advantage of the proposed system is its ability to detect microcalcifications even in mammograms of very dense breasts. A data set of 40 mammograms (Nijmegen database) containing 105 clusters of microcalcifications was studied. Microcalcifications can be accurately and efficiently detected using the proposed approach.