Artificial Intelligence in Colorectal Cancer Screening

Abstract    Purpose of reviewArtificial intelligence (AI) has been shown to be an increasingly powerful tool in the practice of medicine. There is great potential for AI to improve the quality of colorectal cancer screening. We will review recent progress that has been made in this area.Recent findingsThe use of CADe polyp detection has shown potential to increase adenoma detection rate (ADR) and decrease the adenoma miss rate (AMR), while CADx polyp diagnosis has the potential to reduce the need for histopathology, supporting cost-saving strategies like “diagnose and leave” or “resect and discard” during colonoscopy. Computer vision algorithms can measure a variety of colonoscopy quality metrics including bowel preparation adequacy and cecal intubation. Natural language processing is making data abstraction and analytics for colonoscopy qua lity assessment easier and more efficient. Machine learning algorithms also have the potential to personalize screening recommendations for colorectal cancer based on demographics and laboratory blood tests. Although there are barriers to widespread utilization of these new technologies, there is gr eat potential for artificial intelligence to improve the quality of colonoscopy for colorectal cancer screening.SummaryRecent advances in AI have demonstrated great potential for improving the quality of colorectal cancer screening by improving efficiency and outcomes, and in certain cases, offering a path towards decreasing healthc...
Source: Current Treatment Options in Gastroenterology - Category: Gastroenterology Source Type: research