Using ensemble classifier for small bowel ulcer detection in wireless capsule endoscopy images

Abstract

Wireless capsule endoscopy (WCE) has been widely applied in hospitals due to its great advantage that it can directly view the entire small bowel in human body compared with traditional endoscopies and other imaging techniques for gastrointestinal diseases. However, the large number of the images it produced during each test is a great burden for physicians to inspect. To relief the clinicians it is of great importance to develop computer assisted diagnosis system. In this paper, a new computer aided detection scheme aimed for small bowel ulcer detection of WCE images is proposed. This new scheme utilizes an ensemble classifier, which is build upon K nearest neighborhood (KNN), multilayer perceptron (MLP) neural network and support vector machine (SVM), to detect small intestine ulcer WCE images. As far as we know, the combination of multiple classifiers in the field of endoscopic images has never been studied before. Experiments on our present image data show that it is promising to employ the proposed hybrid classifier to recognize the small bowel ulcer WCE images.