Towards Intelligent Systems for Colonoscopy
Jorge Bernal, Fernando Vilariño, Javier Sanchez · InTech eBooks · 2011
Colorectal cancer is one of the leading causes of cancer related deaths. Colorectal cancer’s survival rate depends on the stage in which it is detected, decreasing from rates higher than 95% in the first stages to rates lower than 35% in stages IV and V (Tresca, A. (2010)); hence the importance of detecting it on its early stages by using screening techniques, such as colonoscopy (Hassinger, J.P., Holubar, S.D. et al. (2010)), which is still considered the gold standard for the screening of patients for colon cancers and lesions. Classical focal colonoscopy has been proved to be a successful tool for colon screening, although other tools are being also used for this purpose, such as Virtual Colonoscopy, Computed Tomography Colonoscopy, Chromoendoscopy or Wireless Capsule Video Endoscopy, among others. In this chapter we present tools that can be used to build intelligent systems for colonoscopy. The idea is, by using methods based on computer vision and artificial intelligence, add significant value to the colonoscopy procedure. Intelligent systems are being used to assist in other medical interventions. For instance, we can build systems that can be used to develop the knowledge bases used by expert systems, such as KARDIO (Bratko et al. (1990), which was developed to interpret electrocardiograms. Another example can consist of developing a system that, in the context of anesthesia, provides a robust/reliable control system that could determine the optimal infusion rate of the drugs (muscle relaxant, anesthetic, and analgesic) simultaneously, and titrate each drug in accordance to its effects and interactions. Such a system would be a valuable assistant to the anesthetist in the operating theater. An example of such a system can be found in the work of Nunes et al. (2005). More close to our topic of interest, colonoscopy, we can find many examples of intelligent systems build to assist in cancer detection. Such is the case of breast cancer detection (Wei et al. (2011)) or prostate cancer detection (Viswanath et al. (2011)). The question that arises now is: how can intelligent systems help in colonoscopy? What kind of applications these systems can be built for? In Figure 1 we depict some of the potential areas related to colonoscopy where an intelligent system can play a key role. As shown in Figure 1, we foresee four different areas where an intelligent system can be introduced and add significant value to the colonoscopy procedure: