Medical-AI trial: can AI-assisted colonoscopy increase detection of adenomas (precancerous lesions)? vs. standard (randomized controlled trial) (NCT07066046)
A randomized controlled trial (RCT) evaluating whether AI-assisted colonoscopy raises the adenoma detection rate (ADR) and improves the accuracy of characterizing colorectal lesions (telling benign from malignant), compared with standard colonoscopy. The primary outcome is the proportion of patients with at least one histologically confirmed adenoma (AI vs. control). Target 1,000; recruiting.
Trial overview (primary data)
- StatusRecruiting
- ConditionsColorectal Cancer (CRC), Adenomatous Polyposis, Colorectal Lesions, Colorectal Cancer
- InterventionsPROCEDURE: Colonoscopy
- SponsorInstituto do Cancer do Estado de São Paulo
- Target enrollment1,000 participants
- Period2025-02-01 〜 2026-12-01
Key points
- An RCT of whether AI-assisted colonoscopy raises the adenoma detection rate (ADR) and improves lesion characterization accuracy vs. standard colonoscopy.
- Primary outcome is the proportion of patients with at least one histologically confirmed adenoma (AI vs. control).
- Most colorectal cancers arise from adenomas (precancerous lesions); finding and removing them at colonoscopy aids prevention.
- Small, flat adenomas are easily missed; real-time AI flagging may help detection.
- Target 1,000; recruiting. An evaluation stage, not an establishment of a specific device effectiveness or replacement of the standard.
- CADe supports detection and CADx judges character, with the primary endpoint being patients in whom an adenoma was confirmed.
1Colorectal cancer begins as an adenoma
Most colorectal cancers arise slowly from adenomas, precancerous lesions that are a type of polyp. Finding and removing adenomas at colonoscopy can prevent cancer. But small, flat adenomas are easy to miss, and how many are caught (the adenoma detection rate, ADR) depends on operator skill and concentration. If AI analyzes the image in real time and flags suspicious lesions, it might reduce misses.
2Tested in a randomized controlled trial
This study tests AI-assisted colonoscopy in a randomized controlled trial (RCT). Per the registry summary, it compares AI-assisted with standard colonoscopy to evaluate whether AI can (1) raise the adenoma detection rate (ADR) and (2) improve the accuracy of characterizing colorectal lesions (telling benign from malignant).
The primary outcome is the proportion of patients with at least one adenoma confirmed by histopathology during colonoscopy, compared between the AI and control groups. The target is 1,000 patients, and it is recruiting.
3AI that finds, and AI that distinguishes
AI support in colonoscopy divides into two roles. This trial takes both as objects of evaluation.
Most colorectal cancer arises over time from adenomas, a precancerous lesion. Small flat adenomas are easily missed, and how many are caught depends on the operator's skill and concentration. The primary endpoint is the proportion of patients in whom at least one adenoma was histologically confirmed during colonoscopy, targeting 1,000 cases.
Why it matters
AI support for colonoscopy (CADe/CADx) is one of the most deployed medical-AI areas. Comparing the adenoma detection rate against the standard in an RCT is a useful way to confirm, as evidence, whether AI reduces misses and aids prevention (this trial is at the evaluation stage, not an establishment of effectiveness).
FAQ
What is the adenoma detection rate (ADR)?
Does AI replace the doctor colonoscopy?
Sources (primary)
Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.
- ClinicalTrials.gov (study record, original)
- NCT ID: NCT07066046