Machine learning techniques for bleeding detection in capsule endoscopy

Wireless Capsule Endoscopy (WCE) is a non-invasive procedure to examine the entire gastroenterological tract, especially the small bowel area for any signs of bleeding. Prominent efficient Machine Learning (ML) techniques which are used for the detection of obscure bleeding have been discussed in this paper. The general phases of the procedure include (a) Video acquisition through wireless capsule inhaled by the patient (b) Video frames' pre-processing and feature extraction (c) Classification of bleeding through ML techniques (d) Quantitative performance analysis of the testing results. Major state-of-the-art techniques for bleeding detection in WCE have been studied comprehensively along with their comparative analysis in this survey.