A Region Growing Segmentation for Detection of Microcalcification in Digitized Mammograms

This paper presents an approach for detecting microcalcifications in digital mammograms. The microcalcifications appear i n small clusters of few pixels with relatively high intensity compared with their neighboring pixels. The processing scheme used here focuses on detection of microcalcification in a very weak contrast to their background and presents a computerized technique to identify the microcalcification masses by extracting various discriminating features using the image processing and analyzing algorithms. Preliminary experiments indicate that further studies are needed to investigate the potential of employing region-based segmentation as a tool for detecting microcalcifications in digital mammograms.