AI-powered colonoscopies improve small polyp detection: Study

Colonoscopies using artificial intelligence increased detection of small polyps — but not colorectal neoplasms or advanced adenomas, which can be indicators of colorectal cancer — and resulted in higher rates of unnecessary polyp removal, according to a study published in Annals of Internal Medicine. 

The study, published Aug. 29, assessed the effectiveness of computer-aided detection devices, such as Medtronic’s GI Genius, in detecting advanced colorectal neoplasias, adenomas, serrated polyps, and nonpolypoid and right-sided lesions during colonoscopies by comparing the results with standard colonoscopies. 

Researchers randomly assigned a total of 3,213 patients to a colonoscopy with or without computer-aided detection devices. Computer-assisted colonoscopy methods led to a 55 percent decrease in the miss rate of adenomas. However, there was no significant difference in the advanced colorectal neoplasia detection rate, the rate of detected advanced colorectal neoplasias and the rate of detected advanced adenomas in computer-aided colonoscopies compared to traditional procedures. 

The results “show that it is necessary to continue improving this technology by using broader image databases to train computer-aided detection devices to recognize these types of lesions, and that there is significant room for improvement for this promising technology,” Carolina Mangas-Sanjuan, MD, PhD, a lead author of the study, told Medscape Medical News on Aug. 28.