CURRENT CONCEPTS IN COMPUTER-AIDED DETECTION FOR CT COLONOGRAPHY

Computer-aided detection (CAD) for CT colonography has been under development for about eight years. It is now sufficiently accurate that clinically significant polyps 1 cm and larger can be reliably detected with 90% sensitivity. The research focus is now shifting to improving sensitivity for detection of smaller polyps, reducing the number of false positive detections, integrating CAD into clinical practice and applying CAD to the setting of fecal tagging and laxative-free bowel preparations. This is a dynamic research field which requires the use of state-of-the-art image processing and machine learning techniques