MedPath

Efficacy of Artificial Intelligence-assisted Colonic Polyp Detection System

Not Applicable
Completed
Conditions
Adenoma Colon
Interventions
Device: AI-assisted colonoscopy
Registration Number
NCT05941689
Lead Sponsor
Xiangya Hospital of Central South University
Brief Summary

This is a randomized controlled multicenter clinical trial of computer-aided detection (CADe) system for the adjuvant diagnosis of intestinal polyps/adenomas ever conducted in a Chinese population. In addition, this study will evaluate the effect of CADe system on adenoma detection of endoscopists under fatigue.

Detailed Description

Not available

Recruitment & Eligibility

Status
COMPLETED
Sex
All
Target Recruitment
1906
Inclusion Criteria

Patients who need to undergo colonoscopy. -

Exclusion Criteria

Patients with CRC, inflammatory bowel disease, previous colonic resection, and antithrombotic therapy precluding polyp resection were excluded.

Study & Design

Study Type
INTERVENTIONAL
Study Design
PARALLEL
Arm && Interventions
GroupInterventionDescription
AI-assisted groupAI-assisted colonoscopySubjects in this group undergo AI-assisted colonoscopy. The AI-assisted system not only has the function of automatic polyp detection, but also has the function of colonoscopy quality control.
Primary Outcome Measures
NameTimeMethod
adenoma detection rateup to 2months

The proportion of patients with at least one histologically proven adenoma or carcinoma

withdrawal time compliance rateup to 2months

The proportion of patients with clean withdrawal time \>6min.

Secondary Outcome Measures
NameTimeMethod
adenoma per colonoscopyup to 2 months

The number of adenomas detected per colonoscopy

polyp per colonoscopyup to 2 months

The number of polys detected per colonoscopy

polyp detection rateup to 2 months

The proportion of patients with at least one polyp.

Trial Locations

Locations (2)

Xiangya Hospital Central South University

🇨🇳

Changsha, Hunan, China

Loudi Central Hospital

🇨🇳

Loudi, Hunan, China

© Copyright 2025. All Rights Reserved by MedPath