Artificial intelligence (AI) technologies to help detect or characterise colorectal polyps

Polyps are small growths in the bowel. It is important to detect and identify these because some may develop into bowel (colorectal) cancer if left untreated. Colonoscopy is a procedure that allows healthcare professionals to look inside the bowel using a thin, flexible tube with a camera to check for problems such as polyps or early signs of cancer. Colonoscopy is usually done as an outpatient procedure, with the option of sedation or pain relief.

Artificial intelligence (AI) technologies support polyp detection during colonoscopy by highlighting areas of concern for the endoscopist in a process known as computer-aided detection (CADe). Some AI technologies also offer computer-aided diagnosis (CADx), which helps assess polyp features to guide decisions about removal. Ten AI technologies to assist colonoscopy in the detection of polyps were identified as relevant to the NHS (table 1). All 10 technologies featured CADe and 4 also featured CADx. The cost of the technologies is presented in table 2 of the external assessment group report.