Automatic detection of microcalcifications in digital mammograms by entropy thresholding

A system for automatically detecting microcalcifications in digital mammograms is presented. The proposed system, based on a sequence of preprocessing steps (gradient enhancement, mean contrast enhancement and Gaussian blurring-deblurring process) followed by segmentation using entropy thresholding, can effectively and efficiently detect the suspicious microcalcifications. The performance of the designed system is evaluated by skilled radiologists and shows encouraging results.