Computer Aided Diagnosis of Colorectal Polyps

The purpose of this study is to assess whether computer aided technology (CAD) can help in the diagnosis of polyps found the bowel compared with visual inspection alone and therefore whether it is beneficial in helping clinicians to decide whether to remove a polyp or not. Presently, most endoscopists remove all polyps found and send them to the laboratory for testing. The number of colonoscopies is increasing, meaning that more polyps are detected and removed. This comes at a significant cost to the health service and increases the time taken to complete a colonoscopy.

Removing precancerous polyps from the bowel during a colonoscopy (camera test) is the cornerstone of colorectal cancer screening and prevents polyps developing into bowel cancer. Most polyps develop in the rectosigmoid colon (lower part of the bowel). Many polyps never grow into cancer and it can be difficult for the clinicians performing the procedure (endoscopists) to tell which ones are precancerous. This means many polyps are removed unnecessarily, with a considerable waste of resources. A recent preliminary study indicates a novel artificial intelligence system (EndoBRAIN) for computer-aided diagnosis may be able to distinguish different types of polyps during colonoscopy and therefore help doctors decide which polyps to remove. This study aims to compare the in accuracy of artificial intelligence against the endoscopist's assessment for diagnosis of diminutive (\<5mm) polyps in the lower colon. Patients who are age 18 years or older who undergo colonoscopy for any indication at the participating clinical centres and are diagnosed with diminutive rectosigmoid polyps are eligible for study enrolment. For each detected polyp in the rectosigmoid colon, endoscopists will assess the polyp type using standard colonoscopies (cameras) and then with the use of the EndoBRAIN technology. The polyps will be removed and sent to the laboratory for testing. The difference between clinician diagnosis and EndoBRAIN diagnosis will be compared with the laboratory findings. We hypothesize that the EndoBRAIN technology provides a superior accuracy in identifying precancerous rectosigmoid polyps, compared to endoscopist's own prediction with a standard colonoscope. If the trial confirms the superior accuracy of the EndoBRAIN system, polyps classified as non-cancerous with the EndoBRAIN system no longer need to be removed, meaning a large gain for patients and society, due to significantly less polypectomies and pathology reviews.