FDA Approves First AI Device to Detect Colon Lesions

Laird Harrison

April 09, 2021

The US Food and Drug Administration (FDA) on Friday announced its first-ever approval of an artificial intelligence device to help find colon lesions during colonoscopy.

The GI Genius (Cosmo Artificial Intelligence) identifies areas of the colon where a colorectal polyp or tumor might be located. Clinicians then follow up with a closer examination and possible treatment.

"With the FDA's authorization of this device today, clinicians now have a tool that could help improve their ability to detect gastrointestinal lesions they may have missed otherwise," said Courtney H. Lias, PhD, acting director of the FDA's Gastrorenal, ObGyn, General Hospital, and Urology Devices Office, in a media release.

The GI Genius consists of both hardware and software designed to work with an endoscope. It uses machine learning to recognize possible polyps during a colonoscopy. It marks these areas with green squares on the video generated by the endoscope's camera and emits a short, low-volume sound. Clinicians decide if a lesion is truly present and whether to sample or remove such a lesion.

The GI Genius (Cosmo Artificial Intelligence) identifies specific areas of the colon where a colorectal polyp or tumor might be located.

The device does not diagnose the lesions or recommend treatments and is not intended to take the place of laboratory sampling

The FDA based its approval on a trial in which 700 people 40-80 years of age underwent colonoscopies for colorectal cancer screening, surveillance, follow-up from positive results of a fecal occult blood test, or gastrointestinal symptoms of possible colon cancer.

Of these participants, 263 were being screened or surveilled every 3 years or more. The researchers randomly divided them into a group of 136 who underwent white-light standard colonoscopy with the GI Genius, and 127 who underwent white-light standard colonoscopy without the GI Genius.

Using the GI Genius, clinicians identified adenomas or carcinomas that were later confirmed through lab results in 55.1% of patients. Without the GI Genius, the clinicians identified such lesions in 42.0% of patients.

The patients examined with the GI Genius received more biopsies, including slightly more that were not adenomas. But the biopsies did not lead to any adverse events such as perforations, infections, bleeding, or further biopsies.

More information on the GI Genius is available on the FDA website.

Laird Harrison writes about science, health and culture. His work has appeared in national magazines, in newspapers, on public radio and on websites. He is at work on a novel about alternate realities in physics. Harrison teaches writing at the Writers Grotto. Visit him at www.lairdharrison.com or follow him on Twitter: @LairdH.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.