What is this study trying to find out about colorectal adenoma?

This study aims to evaluate an AI-assisted system designed to support doctors in diagnosing colorectal tubular adenomas. Researchers are focusing on how well the AI can detect adenomas and predict disease risk during colonoscopy. The study is led by Renmin Hospital of Wuhan University and explores new ways to improve diagnosis using advanced imaging and AI technology.

Who may be eligible for this colorectal adenoma clinical trial?

Adults aged 18 years or older who need a colonoscopy can be eligible regardless of gender. Participants must be willing to sign consent and follow study procedures. People with recent abdominal surgery, active lower gastrointestinal bleeding, or pregnancy are excluded. Please review eligibility criteria for more details.

Is there any cost to participate in this colorectal adenoma clinical trial?

Study-related care, including colonoscopy procedures and the AI-assisted system evaluation, is provided at no cost. Personal expenses such as travel or accommodations may vary and are not guaranteed to be covered. Participants should plan for their own travel costs.

Actively Recruiting

Age: 18Years +

All Genders

ID07073430

Study of AI-Assisted System for Diagnosing Colorectal Tubular Adenomas Using Colonoscopy Images

Led by Renmin Hospital of Wuhan University · Updated on 2026-03-25

4000

Participants Needed

Research Sites

N/A

Total Duration

AI-Summary

What this Trial Is About

Researchers are conducting a prospective, multi-center observational study to develop and evaluate an artificial intelligence AI-assisted support system for diagnosing colorectal tubular adenomas. The study aims to build a comprehensive trinity database combining white light, magnifying chromo, and pathological images to simulate doctors decision-making processes. This approach focuses on creating a multimodal deep learning diagnostic model and an interpretable risk prediction model for intestinal adenomas, addressing limitations in previous AI imaging models related to explainability. Participants will undergo colonoscopy examinations in two groups a traditional colonoscopy group and an AI-assisted colonoscopy group. The AI system detects and marks polyp positions on high-definition monitors during procedures, using hollow blue boxes for polyps and hollow red boxes for adenomas. The diagnostic model is based on multimodal endoscopic LAFEQ methods combined with Narrow Band Imaging NBI to enhance detection accuracy. During the study, researchers will measure the accuracy of adenoma diagnosis and the AI systems ability to predict disease risk levels during endoscopy. Participants will be monitored throughout the colonoscopy procedures, with data collected from video recordings and imaging. The study seeks to improve diagnostic precision while providing clear explanations for AI-based decisions, with participation lasting through the colonoscopy process and associated evaluations.

CONDITIONS

Brief Title

Application Evaluation Research on the Artificial Intelligence-assisted Support System for the Diagnosis of Colorectal Tubular Adenoma Lesions

Who Can Participate

Age: 18Years +

All Genders

Eligibility Criteria

You may qualify if you...

Patients aged 65 18 years who need to undergo colonoscopy, regardless of gender
Voluntarily sign the informed consent form
Promise to follow the research procedures and cooperate throughout the study

You will not qualify if you...

History of abdominal or pelvic surgery or radiotherapy in the past
Definite active lower gastrointestinal bleeding
Existing or suspected hereditary colorectal polyposis or inflammatory bowel disease
Uncontrolled hypertension (systolic > 160 mmHg or diastolic > 95 mmHg after treatment)
History of stroke, coronary artery disease, or vascular disease
Pregnant
Unable to complete intestinal preparation

Your Study Journey

Screening

Duration - 2 to 4 weeks

Participants are screened for eligibility to participate in the trial.

1 visit (in-person)

Implementation

Duration - Day of colonoscopy procedure

Participants undergo colonoscopy examinations either with traditional methods or with an AI-assisted support system to detect colorectal tubular adenoma lesions.

1 visit (in-person)