Medtronic's GI Genius intelligent endoscopy module is being evaluated in the National study of Artificial Intelligence in Adenoma Detection for colonoscopy (NAIAD), a large-scale clinical trial in the UK. The AI-powered tool is designed to enhance the detection of precancerous lesions during colonoscopy screenings, addressing the critical need for early detection in colorectal cancer.
The Need for Enhanced Detection
Colorectal cancer is a significant health concern, causing over 150,000 deaths annually in Europe. Early detection dramatically improves outcomes, with a 90% recovery rate when the cancer is found at an early stage. However, initial polyps can be small and easily missed during standard endoscopy, even by experienced endoscopists. This challenge is compounded by increasing workloads in the specialty.
How GI Genius Works
GI Genius acts as a "second set of eyes" for gastroenterologists, reviewing colonoscopy images in real-time and highlighting suspicious regions. The system provides an overlay marker and AI assessment, categorizing areas as adenoma, non-adenoma, or no prediction. While the final decision rests with the endoscopist, GI Genius aims to increase diagnostic confidence and maintain vigilance throughout the procedure, supporting the detection of colorectal polyps of various shapes, sizes, locations, and histology.
Clinical Evidence and Ongoing Research
GI Genius is the most studied AI system in endoscopy. Prior research demonstrated a 14.4% increase in adenoma detection rate and approximately a 50% reduction in adenoma miss rate with the system's use. Smaller lesions (≤ 5 mm) detection improved significantly. The system also helps maintain consistent care by boosting diagnostic confidence among less experienced examiners.
Two major research initiatives are underway to further assess GI Genius. The COLO-DETECT trial, a randomized controlled study involving 2,000 participants across 11 UK sites, found that GI Genius significantly increased adenoma detection rate and increased sessile serrate polyp detection, leading to a recommendation for its use in all colonoscopies. The NAIAD study, supported by NHS England, will evaluate the system's diagnostic performance and impact on endoscopists in a real-world clinical setting. Led by King's College London, the trial will deploy 60 AI systems across 20 NHS hospitals in England over two years.
Expert Opinion
Dr. Bu Hayee, Consultant Gastroenterologist and Principal Investigator of the NAIAD study at King's College Hospital, stated, "Our NAIAD study is set to explore the use of AI in a 'real world' setting, and how it might influence endoscopists in their day-to-day practice."
Future Implications
The use of AI in endoscopies is expected to expand, with GI Genius having the potential to host multiple real-time applications. This advancement promises to improve patient outcomes and enhance the cost-effectiveness of colonoscopies through accurate optical diagnosis, potentially reducing the need for histopathological analysis of every suspicious finding.
