Researchers investigating a novel device that utilizes artificial intelligence (AI) to improve detection of colonic polyps found that the device increased the adenoma detection rate (ADR) and the adenomas per colonoscopy (APR), according to a poster presentation at the American College of Gastroenterology (ACG) 2020 clinical meeting and postgraduate course.

The poster, presented by Aasma Shaukat, MD, from the University of Minnesota, received the ACG Governors Award for Excellence in Clinical Research (Physician, Nontrainee) and the Presidential Poster Award.

In “Improvement in Adenoma Detection Using a Novel Artificial Intelligence-Aided Polyp Detection Device,” the team reported that their work prospectively evaluated the performance of an AI-aided polyp detection device (APDD) in improving ADR and APC, and in detecting sessile serrated lesions (SSLs) during colonoscopies.

The researchers compared ADR, APC, and detection of SSLs in 28 screening and 55 surveillance colonoscopies which included use of a real-time APDD (Iterative Scopes, Cambridge MA) with the outcomes of 117 screening and 166 surveillance colonoscopies among 283 historical controls. They found that the ADR with APDD was 54.2% in all colonoscopies; without APDD the ADR was 40.6% (p=0.028). Among screening colonoscopies, the ADR was 53.6% with APDD and 30.8% without APDD; for surveillance colonoscopies, the rates were 54.5% with APDD and 47.6% without APDD.

The rate of overall SSLs per colonoscopy (SSLPC) with APDD was 0.24 while the rate without APDD was 0.14. (p=0.178). The true histology rate (THR) with APDD was 73.8%, and without APDD the rate was 78.4% (p=0.463); for screening colonoscopies the rates with and without APDD were 75.0% and 71.0% respectively (p=0.731).

“A novel APDD increased the ADR and APC in a cohort of patients undergoing screening and surveillance colonoscopy without concomitant increase in hyperplastic polyp resection,” Dr Shauka said. “Additionally, there was improvement in detection of SSLs in screening patients. AI-aided colonoscopy has the potential for improving the outcomes of patients undergoing colonoscopy.”

—Rebecca Mashaw

Reference:

Shaukat A, Colucci D, Erisson L, et al. Improvement in adenoma detection using a novel artificial intelligence-aided polyp detection device. Talk presented at: American College of Gastroenterology 2020 Clinical Meeting. October 24, 2020. Virtual