Segmentation of microcalcification regions in digital mammograms using self-guided region-growing

摘要

This paper presents a novel segmentation technique to diagnose the microcalcifications in digital mammogram images using the principle of an intuitive region-growing and a simple image enhancement technique named as histogram expansion. This technique uses the intensity values of the enhanced mammogram as the guideline and segments the regions for subsequent diagnosis of microcalcifications. This technique was tested on fifty different mammograms of different orientations. The performance of this technique is proved to be consistently precise and accurate on all those images. The outcome of this mechanism can suitably be subjected to image analysis to study the severity of breast cancer.