Automatic Segmentation of Micro-calcification Based on SIFT in Mammograms

Manual segmentation of micro-calcifications in mammogram can provide clinicians with useful information, such as an estimation of the quantification and the size of abnormalities. However, it is a time and labour consuming process. Automatic segmentation has the potential to assist both in the diagnosis of the disease and in treatment planning. This paper presents a novel mammogram image segmentation algorithm that makes use of Scale Invariant Feature Transform (SIFT) to compute the key point in the suspicious area of the mammograms. A database from MI AS is used in this approach. Initial results are presented to show that SIFT can be used to by computing the key-points to segment micro-calcifications of the mammograms. Further work will focus on finding the ways to set the threshold of the segmentation automatically.