In a recent study published within the journal EClinicalMedicine, researchers in China conducted a scientific review followed by a meta-analysis of studies describing the usage of artificial intelligence (AI) based methods to detect colorectal neoplasia during colonoscopies to know its success in enhancing adenoma detection rates and lowering adenoma miss rates.

Study: Artificial intelligence for colorectal neoplasia detection during colonoscopy: a scientific review and meta-analysis of randomized clinical trials. Image Credit: Peter Porrini / Shutterstock

Background

Colorectal cancer is considered one of the three most prevalent types of cancer across the globe and a considerable contributor to cancer-related mortality rates. The early detection of adenomatous polyps through colonoscopies and their removal is considered one of the chief methods to scale back the incidence of colorectal cancer. While an adenoma detection rate increase of 1% often correlates to a 3% lower risk of colorectal cancer, the variation across endoscopy services also leads to an roughly 27% adenoma miss rate on account of cognitive or technical limitations.

Artificial intelligence-based methods have been extensively explored in recent times to standardize the detection of polyps during colonoscopies to bypass human error. Nonetheless, aside from inconsistent results from studies examining the usage of AI-based adenoma detection tools, concerns also exist concerning the overdiagnoses of polyps, resulting in unwarranted patient burden and costs. Moreover, potential problems with endoscopic training and endoscopist distraction also present challenges.

In regards to the study

In the current study, the researchers conducted an intensive seek for randomized controlled trials evaluating the benefits and drawbacks of using AI-based systems for detecting adenomas and comparing them to plain colonoscopy-based detection methods. This comprehensive review and subsequent meta-analysis aimed to enhance our understanding of AI-based detection methods for colorectal neoplasia.

Studies were included within the systematic review if the colonoscopy being conducted within the enrolled participants was for primary screening, symptoms, or surveillance, and the studies compared AI-based colonoscopy methods with conventional colonoscopy methods. Moreover, only randomized controlled trials that reported outcomes relevant to this study were included. Studies involving patients with hereditary polyposis syndromes or inflammatory bowel disease were excluded.

The outcomes of interest were adenoma detection rate, adenoma miss rate, and adenomas detected in each colonoscopy. These primary outcomes were also stratified by morphology, pathology, location, and size. The secondary outcomes of interest were polyp detection rate, polyp missed rate, procedure time, false alarms, antagonistic events, and the variety of polyps detected in each colonoscopy.

Data extracted from the studies included characteristics of the patients, the study, intervention, detected polyps and adenomas, and the first and secondary outcomes. For tandem trials, only the primary colonoscopy data was used for the meta-analysis to stop the carryover effect. The heterogeneity between studies was quantified using prediction interval, and subgroup analyses and meta-regression were conducted to know the potential heterogeneity sources.

Results

The outcomes showed that the usage of AI-based colonoscopy methods resulted in significant enhancement of colorectal neoplasia detection and substantially lowered the adenoma miss rate and polyp miss rate. The studies using AI-enabled colonoscopy also reported a big elevation in polyp detection rates and adenoma detection rates and within the variety of adenomas and polyps detected during each colonoscopy.

The polyp miss rate from AI-based colonoscopy methods was 52.5% lower, while the polyp detection rate was found to be 23.8% higher. Compared to standard colonoscopy methods, the variety of polyps detected per colonoscopy was 0.271 higher. Nonetheless, the studies showed substantial heterogeneity with regard to the outcomes related to polyp detection.

Similarly, the adenoma detection rate and the adenoma miss rate showed a rise of 24.2% and a decrease of fifty.5%, respectively, when employing AI-based colonoscopy methods. Moreover, about 0.202 more adenomas were detected per colonoscopy using AI-enabled adenoma detection methods. Nonetheless, much like polyp detection outcomes, the outcomes from the randomized controlled trials also showed significant heterogeneity in the outcomes.

Conclusions

Overall, the findings suggested that using AI-enabled colonoscopy could significantly improve the detection of adenomas and colorectal neoplasia. Moreover, slight improvements in the standard of colonoscopies could translate to potential net gains in large-scale screening programs for colorectal cancer while maintaining the homogeneity and quality of colonoscopies.

The researchers also discussed future research implications of those findings, including the necessity for longitudinal studies to verify the efficacy of AI-based colonoscopic adenoma detection methods in lowering the morbidity and mortality related to colorectal cancer.