Accuracy of Real Time Characterization in Artificial Intelligence-assisted Colonoscopy

Colorectal cancer (CRC) is the third most common cancer, and the second most common cause of cancer-related death worldwide. CRC screening is used for detection and removal of precancerous lesions before they develop into cancer. Colonoscopy is regarded being superior to other screening tests, and is therefore used as the golden standard. Screening colonoscopy is associated with a reduced risk of CRC-related death. Since it is not possible for an endoscopist to determine the histopathology of the polyp with certainty during a colonoscopy, detected pre-malignant lesions should be removed and sent for histological examination. Multiple studies have shown that there is a strong association between findings at the baseline screening colonoscopy and rate of serious lesions at the follow up colonoscopy. Risk factors for adenoma, advanced adenoma and cancer at follow-up colonoscopy are multiplicity, size, villousness, and high degree dysplasia of the adenomas at the baseline screening colonoscopy. Within the last few years there have been published several randomized controlled trials (RCT) investigating the efficacy of real time computer-aided detection. Studies have shown that AI contributes to a significantly higher adenoma detection rate (ADR), compared colonoscopies without assistance of an AI system.There have been concerns about prolonged colonoscopy time, and increased workload if implementing the AI-system, since the increased detection of small polyps may lead to unnecessary polypectomy. With the development of computer-aided polyp characterization (CADx) systems, it is possible to use AI for decision support and not only for detection. There is no evidence yet that the CADx system increases the sensitivity for small neoplastic polyps when used by non-expert endoscopists (accredited for standard colonoscopy), but it may improve the clinicians confidence, and increase the specificity for optical diagnosis (Barua et al). Diminutive polyps (1-5 mm) in the rectosigmoid colon can be left in situ when diagnosed with high confidence with a sensitivity of at least 90% and a specificity of at least 80%. To implement the resect-and-discard strategy, a sensitivity of at least 80% is acceptable. This is recommended by the European Society of Gastrointestinal Endoscopy (ESGE) as a strategy to decrease the unnecessary removal of small polyps with a negligible risk of harbouring cancer. Although the resect-and-discard strategy is assessed to be a safe and cost-effective method, it is important to be cautious with lesions in the right colon due to their malignant potential. Reliable CADx systems could enable a more targeted removal of neoplastic polyps, while diminutive non-neoplastic polyps could be left behind. The potential excessive workload due to the CADe system could therefore theoretically be avoided by adding the CADx system. The results so far are promising, suggesting that AI-assisted colonoscopy is superior to conventional colonoscopy when it comes to polyp and adenoma detection. Continued improvement of CADx systems in differentiating the pathology of colorectal lesions is needed, as well as additional clinical studies to assess the potential value of the CADx system. The overall aim of this research is to investigate the quality, and the possible benefits of AI-assistance in colonoscopy. Hopefully this can contribute to a more accurate, safe, and targeted diagnosis and treatment of patients in the future. The investigators have designed a quality assurance study to investigate the effect of real time AI-assisted colonoscopy with the CADx system (GI Genius, Medtronic). This study "REG-093-2022" is a substudy to the RCT "REG-092-2022". The investigators wish to evaluate the diagnostic accuracy of the CADx system.