chapter  19
Computer Aided Diagnosis in Mammography with Emphasis on Automatic Detection of Microcalcifications
WithYong Man Ro, Sung Ho Jin, Konstantinos N. Plataniotis
Pages 32

Breast cancer is one of the major causes of increasing mortality in women, especially in developed countries including the USA and Europe. The importance of automatic diagnosis of microcalcifications in early detection and diagnosis procedures led to the development of computer aided diagnosis (CAD) in mammography. Digital database for screening mammography (DDSM) is obtained by scanning the mammography film. Full-field digital mammography (FFDM) utilizes solid-state detectors, similar to those found in digital camera, that convert X-rays into electrical signals producing images with better signal-to-noise ratio (SNR) compared to DDSM images. The chapter provides information on technologies and implementation details regarding the background segmentation, contrast enhancement and denoising steps. It details support vector machine (SVM)-based detection systems for microcalcifications and associated detail algorithms. The chapter presents experimental results to show the demonstration of microcalcification enhancement and microcalcification detection.