Region labeling of colonoscopic images using fuzzy logic

A fuzzy rule base approach to the labeling of colonoscopic images to render assistance to the clinician is presented. The color images are segmented using a scale-space filter. Several features are selected and fuzzified. The knowledge-based fuzzy rule base system performs the labeling of the segmented regions into background, lumen, and abnormalities (polyps, bleeding lesions). The system has been tested on several precaptured colonoscopic images. The results are encouraging. Modifications are in progress to prepare for real-time clinical testing.