Polyp Classification Based on Deep Neural Network for Colonoscopic Images

In this paper, a colorectal polyp classification method based on deep neural network (DNN) was proposed for BLI (Blue Laser Imaging) images. Since polyps can be considered as objects, an one-stage object detection network, YOLO (You Only Look Once), is selected to develop a computer-aided system to detect and classify polyps. Based on data augmentation and transfer learning, the DNN was modified to classify polyps into two classes: hyperplastic and adenomatous. To evaluate the performance of the proposed method, many colonoscopic images are collected for testing. The precision and recall rates can achieve 99% for 234 cases outside the training set. Experimental results show that the proposed method can not only detect but also classify colorectal Polyps in BLI images.