Computer-aided diagnostic system for the breast cancer by detecting microcalcifications

Abstract

This thesis presents a CAD (Computer-Aided Diagnosis) system for detection of the breast cancer using digitized X-ray mammograms. The system is composed of personal computer, X-ray film scanner, high-resolution display and application software. In order to diagnose the breast cancer, detection and classification method of microcalcifications is developed, which are the one of early sign of the breast cancer. Image enhancement algorithm is also implemented in this system. To verify the performance of the proposed methods, the comparison study was performed on the enhancement algorithms, which are derived from already proposed one. Adaptive selection method of the threshold value using local statistics is proposed for extracting the surrounding region dependence method (SRDM) matrix instead of the fixed value selection. Its performance is better than that of the fixed threshold value.