Bilal Taha, Naoufel Werghi, and Jorge Dias
Polyp detection, Endoscopy videos, Deep learning, Texture features, Shape features
Early detection of polyps play an essential role for the prevention of colorectal cancer. Manual clinical inspection have many limitations and could result to either false or missed polyps. Computer aided diagnosis system has been used to help the medical expert and to provide more accurate diagnosis. Since their introduction, many types of algorithms have been proposed in the literature using different types of features and classifiers. This paper provides a state-of-the-art for the automatic detection of polyps using endoscopic videos. Given the increasing evolution of medical imaging technologies and algorithms, it is important to have a recent review in order to know the current state of the art, and the opportunities for improving existing algorithms, or developing innovative ones. The paper divides the work done on this research area according to the type of features and classification methods implemented. The features have been divided into shape, texture or fusion features. Future directions and challenges for more accurate polyp detection in endoscopy videos are also discussed.