Extracting hidden visual information from mammography images using conjugate image enhancement software

Abstract

Most early breast cancers can be diagnosed by detecting calcification clusters in mammography X-ray images. The clusters appear as groups of small, bright particles with arbitrary shapes. Detecting micro-calcifications is difficult because they are embedded in a non-homogeneous background. Many missed radiology diagnoses can be attributed to human factors such as the use of subjective criteria or variable criteria in decision making, distraction by other image features, the large number of images to be inspected, or just simple oversight. Consequently there are very good reasons for pursuing reliable and effective methods for micro-calcifications detection. While many methods for micro-calcification segmentation have been developed in the past ten years, they either require manual threshold adjustments or depend on local statistics to compute those thresholds. This paper presents a new fully automated, parameter-free, and local statistics independent, algorithm for micro-calcification segmentation in mammography X-ray images.