Review of Computer-aided detection of breast cancer on mammograms

Breast cancer is the most common form of cancer among women worldwide. Early detection of breast cancer can increase treatment options and patients survivability. Recently, computer-aided detection (CAD) systems have been developed to use by less experienced radiologist and to reduce the number of false-negative interpretation caused by radiologist distraction and complex architecture. Numbers of Recent researches and FDA-approved commercially CAD system were evaluated. This review is based on breast density, lesions and their stage using image processing, pattern recognition, and artificial intelligence techniques based on the knowledge from too many mammograms which are had been interpreted. Early detecting Breast cancer in single reading with CAD has better performance compared with double reading, although false positive mark increased in dense breast that has a lower false than interpretation without CAD. By assessment of different CAD techniques, other modalities like ultrasound imaging are suggested to achieve additional information. One solution to improve CAD systems is the development of interactive CAD systems that can process only regions of interest. With using more database and large number of data we can improve these CAD systems and helps the radiologist in detecting the abnormalities in an efficient way in early detecting Breast cancer.

Keywords:

Breast Cancer , Computer aided systems (CAD)