AI accurately IDs diminutive polyps during colonoscopy

Computer-aided diagnosis (CAD) powered by artificial intelligence can accurately assess diminutive colorectal polyps, according to a new study published in Annals of Internal Medicine. But is the CAD’s performance level high enough that specialists can follow the recommended “diagnose-and leave” strategy for diminutive polyps?

The authors studied data from more than 700 consecutive patients who underwent colonoscopy at a single facility in Japan from June to December 2017. Overall, 466 diminutive polyps from 325 patients were assessed by CAD. The success rate for acquiring endocytoscopic images was 100 percent. The success rate for pathologic prediction by CAD was more than 98 percent.

The authors wanted to see if the CAD would achieve a NPV of 90 percent or greater, since that would mean it could sufficiently be used by specialists adopting the diagnose-and-leave strategy.

Overall, the CAD-stained analysis had a negative predictive value (NPV) of more than 93 percent in the worst-case scenario, which assumed that polyps lacking CAD diagnosis or pathology were all true- or false-negative, and more than 96 percent in the best-case scenario, which assumed that polyps lacking CAD diagnosis or pathology were all true- or false-positive.

“Real-time use of the fully automated CAD system designed for endocytoscopes can meet the clinical threshold required for the diagnose-and leave strategy for diminutive, nonneoplastic rectosigmoid polyps, which may help improve the cost-effectiveness of colonoscopy,” wrote author Yuichi Mori, MD, of Japan’s Showa University Northern Yokohama Hospital.

The authors added that the NPVs for diminutive rectosigmoid adenomas were 96.4 percent for CAD, 91.8 percent for two expert endoscopists and 86.6 percent for two nonexpert endoscopists.

Michael Walter
Michael Walter, Managing Editor

Michael has more than 19 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.

Around the web

The new F-18 flurpiridaz radiotracer is expected to help drive cardiac PET growth, but it requires waiting between rest and stress scans. Software from MultiFunctional Imaging can help care teams combat that problem.

News of an incident is a stark reminder that healthcare workers and patients aren’t the only ones who need to be aware around MRI suites.

The ACR hopes these changes, including the addition of diagnostic performance feedback, will help reduce the number of patients with incidental nodules lost to follow-up each year.