Computer-aided detection tied to higher detection rates of colorectal neoplasia: Lancet

Italy: Gastroenterologists can benefit from computer-aided detection (CADe) approaches based on artificial intelligence algorithms in identifying colorectal neoplasia.

According to a new research, detection rates of colorectal neoplasia are greater with computer-aided detection (CADe) than with other techniques such as chromoendoscopy or instruments that improve mucosal visibility. This implies that CADe strategies should be more widely included in the community endoscopy services. 

This study was conducted by Marco Spadaccini and the team with the idea of exploring the key role and utility of CADe and its comparison with existing tools in the same field. The findings of this study were published on 1st October 2021 in The Lancet Gastroenterology & Hepatology journal.

Researchers conducted a thorough search of the PubMed/Medline, Embase, and Scopus databases from inception to November 30, 2020, for randomized controlled trials investigating the efficacy of the following endoscopic techniques in detecting colorectal neoplasia: CADe, high definition (HD) white-light endoscopy, chromoendoscopy, or add-on devices (ie, systems that increase mucosal visualization, such as full-spectrum endoscopy [FUSE] or G-EYE balloon endoscopy). Adenoma detection rates, sessile serrated lesion detection rates, the fraction of big adenomas identified per colonoscopy, and withdrawal times were also gathered. The researchers used a frequentist framework and random-effects network meta-analysis to compare artificial intelligence to chromoendoscopy, enhanced mucosal visualization technologies, and HD white-light endoscopy (the control group).

The primary analysis comprised 50 randomized controlled studies with a total of 34 445 individuals. In all 50 investigations, HD white-light endoscopy was used as the control method. When compared to the control approach, the adenoma detection rate with CADe was 74% higher, 44% higher with chromoendoscopy, and 41% higher with enhanced mucosal visibility systems. Cross-comparisons of CADe with other imaging techniques revealed a substantial increase in the adenoma detection rate with CADe against enhanced mucosal visibility systems, as well as with CADe compared chromoendoscopy. As focused on large adenomas (10 mm), there was a substantial increase in detection only with CADe when compared to HD white-light endoscopy; CADe was regarded as the superior method for detection of large adenomas. Although no significant improvement in sessile serrated lesion detection rate was seen, CADe appeared to be the best method for detection of sessile serrated lesions. There was no statistically significant difference in withdrawal time between CADe and the other methods.

"CADe appears to have an advantage over other advanced endoscopy imaging techniques, such as chromoendoscopy or mucosal visualization tools, for increasing the detection of colorectal neoplasia," the authors wrote. "This supports wider incorporation of CADe technologies into community endoscopy services."

Reference:

Spadaccini M, Iannone A, Maselli R, et al. Computer-aided detection versus advanced imaging for detection of colorectal neoplasia: A systematic review and network meta-analysis. Lancet Gastroenterol Hepatol. Published online August 4, 2021.

doi:10.1016/S2468-1253(21)00215-6