Artificial intelligence–assisted real-time monitoring of effective withdrawal time during colonoscopy: a novel quality marker of colonoscopy

Abstract

Background and aimsThe importance of withdrawal time during colonoscopy cannot be overstated in mitigating the risk of missed lesions and post-colonoscopy colorectal cancer. We evaluated a novel colonoscopy quality metric called the effective withdrawal time (EWT), which is an artificial intelligence (AI) derived quantitative measure of quality withdrawal time and their association with various colonic lesions detection rates, as compared with standard withdrawal time (SWT). MethodsA total of 350 video recordings of colonoscopy withdrawal (from caecum to anus) were assessed by the new AI model. The primary outcome was adenoma detection rate (ADR) according to different quintiles of EWT. Multivariate logistic regression, adjusting for baseline covariates, was used to determine the adjusted odd ratios (OR) for EWT on lesion detection rates, with the lowest quintile as reference. The area under the receiver operating characteristic curve (AUC) of EWT was compared with SWT. ResultsThe crude ADR in different quintiles of EWT, from lowest to highest, was 10.0%, 31.4%, 33.3%, 53.5%, and 85.7%, respectively. The ORs of detecting adenoma and polyp were significantly higher in all top 4th quintiles when compared to the lowest quintile. Each minute increase in EWT was associated with a 49% increase in ADR (aOR, 1.49; 95% CI, 1.36 to 1.65). The AUC of EWT was also significantly higher than SWT on adenoma detection (0.80 [95%CI: 0.75 – 0.84] vs 0.70 [95%CI: 0.64-0.74], p <0.01). ConclusionsThe AI-derived EWT monitoring is a promising novel quality indicator for colonoscopy, which is more associated with ADR than SWT.