AI-assisted colonoscopy to find and classify colorectal lesions

Artificial Intelligence-assisted Colonoscopy in the Detection and Characterization of Colorectal Lesions: Randomized Controlled Clinical Trial

This trial tries to see if using AI during routine colonoscopy helps doctors find more adenomas and better classify colorectal lesions in adults.

Quick facts

• Phase: NA
• Study type: Interventional
• Enrollment: 1000 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: Instituto do Cancer do Estado de São Paulo (other)
• Locations: 1 site (São Paulo, São Paulo)
• Trial ID: NCT07066046 on ClinicalTrials.gov

What this trial studies

This randomized controlled trial compares AI-assisted colonoscopy with standard colonoscopy to measure adenoma detection rate and real-time lesion characterization accuracy. Adults scheduled for elective colonoscopy at Hospital das Clínicas da Faculdade de Medicina da USP are randomized to either AI assistance or usual care, with outcomes recorded by endoscopists and pathology confirmation. Key eligibility excludes prior colorectal cancer, inflammatory bowel disease, polyposis syndromes, recent colorectal surgery, emergency procedures, poor bowel prep, or severe comorbidity (ASA ≥3). Primary endpoints are adenoma detection rate and characterization accuracy, with complete colonoscopy to the cecum and adequate bowel preparation required.

Who should consider this trial

Good fit: Adults aged 18 or older who are scheduled for an elective colonoscopy, can give informed consent, and do not have prior colorectal cancer, inflammatory bowel disease, polyposis, recent colorectal surgery, or severe comorbidity are ideal candidates.

Not a fit: Patients with emergency indications, incomplete or inadequately prepped colonoscopies, a history of colorectal cancer, inflammatory bowel disease, polyposis syndromes, recent colorectal surgery, or ASA class 3 or higher are unlikely to benefit from participation.

Why it matters

Potential benefit: If successful, this approach could help clinicians detect more precancerous adenomas and better identify lesion type during colonoscopy, potentially lowering colorectal cancer rates over time.

How similar studies have performed: Multiple recent trials and meta-analyses have shown that AI-assisted colonoscopy can increase adenoma detection rates and improve lesion characterization, so this approach has growing supportive evidence.

Eligibility criteria

Inclusion Criteria:

* All patients aged 18 years or older, with an elective indication for colonoscopy who sign the informed consent form agreeing to participate in the study.

Exclusion Criteria:

* History of inflammatory bowel disease.
* History of colorectal cancer.
* Personal history of colorectal surgery.
* Contraindication to endoscopic biopsies.
* History of intestinal polyposis syndromes.
* Urgent or emergency cases.
* Presence of severe, decompensated comorbidities, or with a score of 3 or higher according to the American Society of Anesthesiologists (ASA) classification.
* Incomplete colonoscopy that does not reach the cecum.
* Insufficient or inadequate bowel preparation, with a score lower than 6 on the Boston Bowel Preparation Scale.
* Patients who do not agree to participate in the study and do not sign the informed consent form (ICF).

Where this trial is running

São Paulo, São Paulo

• Hospital das Clínicas da Faculdade de Medicina da USP — São Paulo, São Paulo, Brazil (RECRUITING)

Study contacts

• Study coordinator: Márcio Roberto Facanali Júnior

Conditions: Colorectal Cancer, Adenomatous Polyposis, Colorectal Lesions, Adenoma, Artificial Intelligence, Colonoscopy, Colorectal lesions