In partnership with Erlanger, researchers at the University of Tennessee Health Science Center College of Medicine – Chattanooga were part of a major multicenter international study just published in the journal Lancet Digital Health. The study evaluated the effectiveness of a new Artificial Intelligence (AI) system for detecting polyps in screening and surveillance colonoscopy. Arslan Kahloon, MD, Chief of Gastroenterology, is the author of the paper. Giuseppe Pizzorno, PhD, PharmD, associate dean for Research and chief research officer at Erlanger, was the Chattanooga team lead on the study.
Though colonoscopy is considered the gold standard for early detection and removal of precancerous polyps, a substantial number of polyps can be missed during the procedure. Missed lesions are a serious concern, as this can lead to interval colon cancer. It is estimated that at least 50% of all post-colonoscopy colorectal cancers develop from them.
The newly-published study took place over 14 months, from 2020 to 2022, across 10 leading medical centers, with 31 endoscopists and 952 enrolled patients. Erlanger was the lead site for recruiting patients, working in collaboration with other U.S. centers such as Johns Hopkins University, Columbia University Irving Medical Center, and Metro Health Medical Center, as well as centers in Germany, Netherlands, and Israel.
The study validated the effectiveness of a new computer-aided system “MAGENTIQ-COLO™”. Results showed that using this AI guided system increases adenoma detection rate by 7%, and decreases the adenoma miss rate by 19%. These numbers are high enough to have a strong connection to the decrease in colorectal cancer occurrence and patient mortality.
“Erlanger and UT Health Science Center College of Medicine – Chattanooga were the first U.S. site to successfully conduct an off-line pilot trial of this novel device,” Dr. Kahloon said. “This led to the formation of a multicenter consortium among experts at 10 institutions across three continents. Our study led to the FDA approval of ‘MAGENTIQ-COLO’ as a new promising computer-aided detection system. Such AI-capable technologies show immense promise to become standard of care in colonoscopy in the future. I am very grateful to our research teams in Chattanooga, including the team at Institute for Clinical Research, GI lab staff, Steven Kessler, MD, and Laxmi Parsa, MD, who were all instrumental in the success of our study.”
The paper, titled “A novel computer-aided polyp detection system in screening and surveillance colonoscopy: an international multicentre randomised, tandem trial,” is in the March 2024 issue of Lancet Digital Health.