Computer Aided Detection and Diagnosis of Colon Polyps with Morphological and Texture Features

In this paper, we propose a new technique to utilize both the morphological and the texture information of the colon wall for detection of colonic polyps. Firstly this method can quickly identify suspicious patches of the colon wall by employing special local and global geometrical information, different from other methods of utilizing local geometry only. By our edge-detection technology, the growing region of suspected polyps is identified and its internal textures are quantitatively analyzed based on an assumed ellipsoid polyp model. Both the extracted texture and morphological information are then applied to eliminate the false positives from the identified suspicious patches. With all the extracted geometrical, morphological and texture features, this presented computer-aided detection method have demonstrated significant improvement in detection of the colonic polyps for virtual colonoscopy.