Key points are not available for this paper at this time.
During a conventional colonoscopy examination, the endoscopist watches a video monitor that displays the real-time images being captured by a CCD camera at the distal end of the colonoscope. Some researchers believe that a computer can be used to interpret these images and be trained to recognize and correctly locate the position of the lumen. This information will then be processed and used to control and maneuver the colonoscope automatically into the colon. It can also be used to identify abnormalities of the large intestine. This article presents a literature survey that gives a critical review of research in the area of visual control in automated colonoscopy. It is concluded that, generally, the idea of using computer vision to aid in the automation of colonoscopy is meritorious. However, it is important for one to realize the vast differences in physical attributes of the colon. No two colons are similar. Even within the same colon, one section may portray very different characteristics from another. Since technology is not yet at the stage where computers can think in the same way as do human beings, computerized visual control in colonoscopy may not be realized for some time.
Phee et al. (Thu,) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: