Autonomous AI vs AI-assisted human optical diagnosis for colorectal polyps
Official title: Autonomous Artificial Intelligence Versus AI Assisted Human Optical Diagnosis of Colorectal Polyps
Run by Centre hospitalier de l'Université de Montréal (CHUM) · NCT06543862 · Data updated 2026-11-12
A computer-aided AI system is tested to predict the nature of colorectal polyps during colonoscopy and compare its autonomous predictions with those of human doctors using AI assistance.
Quick facts
| Study type | Interventional |
|---|---|
| Ages | 45 Years to 80 Years |
| Sex | All |
| Enrollment | 540 (estimated) |
| Sponsor | Centre hospitalier de l'Université de Montréal (CHUM) Academic / other |
| Locations | 1 site (Montreal, Quebec) |
| Status | Recruiting |
| Started | November 15, 2024 |
| Main results expected | October 2026 |
| Study ends | October 2026 |
| Registry record updated | September 21, 2026 |
| Trial ID | NCT06543862 on ClinicalTrials.gov |
Is this trial a fit for you?
✓Likely a good fit
Individuals undergoing full colonoscopy who have colonic polyps and do not have inflammatory bowel disease, active colitis, coagulopathy, familial polyposis, very poor health (ASA > 3), or require emergency colonoscopy.
✕Probably not a fit
Patients with conditions outside the eligibility criteria or whose polyps are not well characterized by AI in real time may not see a benefit.
A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.
What this trial studies
This study evaluates an AI-powered CADx system that analyzes real-time endoscopy video to predict polyp type. It aims to determine whether autonomous AI predictions can match or improve upon endoscopist-driven, AI-assisted optical diagnoses, potentially reducing the need for pathology. The trial explores cases where the AI and endoscopist disagree to identify suboptimal diagnostic scenarios and refine management of serrated polyps. It is conducted in a single center with inclusion criteria requiring a full colonoscopy and excluding patients with inflammatory bowel disease, coagulopathy, or poor general health. The study focuses on real-time decision support and how best to integrate AI into clinical workflow.
Why it matters
Potential benefit: If successful, patients could benefit from faster, potentially more accurate polyp characterization during colonoscopy, reducing unnecessary pathology and guiding immediate management.
How similar studies have gone: Prior CADx studies show AI can match or complement human optical diagnosis in polyp characterization, though autonomous use with endoscopists is still being explored.
Who can join: the full rules
Show full inclusion / exclusion criteria
Inclusion Criteria: * Indication for full colonoscopy. Exclusion Criteria: * Known inflammatory bowel disease * Active colitis * coagulopathy * familial polyposis syndrome * poor general health, defined as an American Society of Anesthesiologists class \>3 * emergency colonoscopy
Where it's running
-
Ghislaine Ahoua
Montreal, Quebec, Canada Recruiting
Study contacts
- Principal investigator: Daniel von Renteln, MD — University of Montreal Medical Center (CHUM)
- Study coordinator: Daniel von Renteln, MD
- Email: daniel.von.renteln.med@ssss.gouv.qc.ca
- Phone: 514 890-8000
Similar studies
Other studies for the same condition or treatment. Recruiting studies are listed first.
-
Sedation with remimazolam versus propofol for colon polyp removal
This test checks whether remimazolam besylate plus sufentanyl gives more stable blood pressure and breathing than propofol plus sufentanyl for adults 18–60 having colonoscopic polypectomy.
-
Teaching doctors to tell Paris-class polyps apart using real and AI-generated colon images
This trial tests if training doctors with AI-generated colon polyp images helps them learn to classify polyps by the Paris system better than training with real images.
-
Using AI to improve polyp detection during colonoscopy
This study is testing if using artificial intelligence during colonoscopy can help doctors find more polyps in people over 50 and tell which ones are harmless or harmful.
-
Using AI to improve polyp detection during colonoscopy
This study is testing whether using AI technology during colonoscopies can help doctors find more polyps in adults in Singapore.
-
A new technique for removing colonic polyps
This study tests a new method for removing colon polyps to see if it helps doctors get rid of them more completely and reduce the chances of them coming back.
How to take part
- Review the eligibility criteria above with your treating physician.
- Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
- Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.