3d Colonic Polyp Detection in Ct Images Using Template Matching

In this paper we present a computer aided detection (CAD) system for colonic polyp detection in computed tomography (CT) images. Here, the segmentation of the colon region was performed in three steps which are thresholding, labeling and analyzing the shape morphologies using decision rules. After extracting the colon region, searching with 3D semi-spherical polyp template was performed to detect the polyps. Because the polyps can appear in all possible orientations on the colon wall, 3D rotations were performed during the search. The similarity was measured using convolution operation. The performance of the system was evaluated using a test set containing 20 cases having 13 polyps which were marked by expert radiologists. When the results were compared with the reviews of the radiologists, it was seen that the system achieved 92.3% sensitivity with 0.84 FP regions per polyp.