Abstract: Colorectal cancer is a cancer that starts inside the colon or the rectum in large intestine. These cancers is likewise called colon cancer or rectal cancer, depending on wherein they start. Colon and rectal cancer are frequently grouped together because they've many features in common and most colorectal cancers begin as a increase on the internal lining of the colon or rectum called a polyp. A few types of polyps can change into cancer over the several years, but not all polyps end up in cancers. The risk of changing into a most cancers depends at the kind of polyp. Computer-aided detection (CADe) and analysis (CAD) has been a rapidly growing, potential area of research in medical imaging. Machine leaning (ML) plays a crucial role in CAD, because objects such as lesions and organs may not be represented accurately with the aid of an easy equation; as a consequence, medical pattern recognition essentially require “getting to know from examples.” Computed tomography (CT) Colonography or virtual colonoscopy makes use of special x-ray machine to have a look at the large intestine for cancer and growths known as polyps. All through the examination, a small tube is inserted a short distance into the rectum to permit for inflation with air at the same time as CT image of the colon and the rectum are taken. CT technologist determines those images to discover the severity of polyp based on its length. In this survey, we review the different papers and journals in the literature that attempted to address these problems and compare various pre-processing steps, classification and segmentation algorithms, feature set considered, which are used to detect and classify polyp in colon cancer and we also focus on various deep learning algorithms used in similar medical diagnosis and how efficiently it is used to solve problem.
Keywords: Colorectal cancer, Computed tomography (CT) Colonography, polyp, Deep learning Algorithms.