Efficacy of Real-Time Computer-Aided Detection of Colorectal Neoplasia in a Randomized Trial

Although rates of detection of adenomatous lesions (tumors or polyps) and cecal intubation are recommended for use as quality indicators for screening colonoscopy, these measurements have not been validated, and their importance remains uncertain. We used a multivariate Cox proportional-hazards regression model to evaluate the influence of quality indicators for colonoscopy on the risk of interval cancer. Data were collected from 186 endoscopists who were involved in a colonoscopy-based colorectal-cancer screening program involving 45,026 subjects. Interval cancer was defined as colorectal adenocarcinoma that was diagnosed between the time of screening colonoscopy and the scheduled time of surveillance colonoscopy. We derived data on quality indicators for colonoscopy from the screening program's database and data on interval cancers from cancer registries. The primary aim of the study was to assess the association between quality indicators for colonoscopy and the risk of interval cancer. A total of 42 interval colorectal cancers were identified during a period of 188,788 person-years. The endoscopist's rate of detection of adenomas was significantly associated with the risk of interval colorectal cancer (P=0.008), whereas the rate of cecal intubation was not significantly associated with this risk (P=0.50). The hazard ratios for adenoma detection rates of less than 11.0%, 11.0 to 14.9%, and 15.0 to 19.9%, as compared with a rate of 20.0% or higher, were 10.94 (95% confidence interval [CI], 1.37 to 87.01), 10.75 (95% CI, 1.36 to 85.06), and 12.50 (95% CI, 1.51 to 103.43), respectively (P=0.02 for all comparisons). The adenoma detection rate is an independent predictor of the risk of interval colorectal cancer after screening colonoscopy. 2010 Massachusetts Medical Society

New England Journal of Medicine, 370(14), 1298-1306

Objective The effect of colonoscopy on colorectal cancer mortality is limited by several factors, among them a certain miss rate, leading to limited adenoma detection rates (ADRs). We investigated the effect of an automatic polyp detection system based on deep learning on polyp detection rate and ADR. Design In an open, non-blinded trial, consecutive patients were prospectively randomised to undergo diagnostic colonoscopy with or without assistance of a real-time automatic polyp detection system providing a simultaneous visual notice and sound alarm on polyp detection. The primary outcome was ADR. Results Of 1058 patients included, 536 were randomised to standard colonoscopy, and 522 were randomised to colonoscopy with computer-aided diagnosis. The artificial intelligence (AI) system significantly increased ADR (29.1%vs20.3%, p<0.001) and the mean number of adenomas per patient (0.53vs0.31, p<0.001). This was due to a higher number of diminutive adenomas found (185vs102; p<0.001), while there was no statistical difference in larger adenomas (77vs58, p=0.075). In addition, the number of hyperplastic polyps was also significantly increased (114vs52, p<0.001). Conclusions In a low prevalent ADR population, an automatic polyp detection system during colonoscopy resulted in a significant increase in the number of diminutive adenomas detected, as well as an increase in the rate of hyperplastic polyps. The cost–benefit ratio of such effects has to be determined further. Trial registration number ChiCTR-DDD-17012221; Results.