Computer-based Classification of Small Colorectal Polyps

Recent studies have shown that narrow-band imaging (NBI) is a powerful diagnostic tool for the differentiation between neoplastic and non-neoplastic colorectal polyps. The aim of the present study is to develop and evaluate a computer-based method for automated classification of small colorectal polyps on the basis of vascularization features.In a prospective study 207 polyps with a size of 10mm or smaller should be detected by a zoom NBI colonoscopy. These polyps were imaged and subsequently removed for histological analysis. the polyp images were analyzed by a computer algorithm. The proposed computer-based method consists of several steps: preprocessing, vessel segmentation, feature extraction and classification. The results of the automatic classification (sensitivity, specificity, negative predictive value, positive predictive value and accuracy)were compared to those of human observers, who were blinded to the histological gold standard.