Multiregion segmentation of microcalcificationin mammogram images by using Parametric Kernel Graph Cut algorithm

Early detection of breast cancer can be detected through screening mammography. However, the potential abnormality such as microcalcification hardly differentiated by radiologists due to tiny size, which sometimes hidden behind density tissue. Therefore, image segmentation technique is required. This paper proposes use Parametric Kernel Graph Cut Algorithm in segmenting microcalcification. The performances this method were measured based on accuracy, sensitivity, Dice and Jaccard coefficient. All experimental results generated satisfying results, whereby all images produced average 91.67% for coefficient 84.72% Meanwhile, both accuracy sensitivity acquired 97.84% 96%, respectively. algorithm had proved its ability segment robustly efficiently.