NCT05444166

Explore the Relationship Between the Percentage of Colonoscopy Withdrawal Overspeed and the ADR

Explore the Relationship Between the Percentage of Colonoscopy Withdrawal Overspeed and the Adenoma Detection Rate: a Prospective, Multicenter, Observational Study

• Status: Unknown
• Phase: —
• Study type: Observational
• Enrollment: 550 (estimated)
• Sponsor: Renmin Hospital of Wuhan University · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

In this study, the investigators used the optical flow method to measure the colonoscopy withdrawal speed, and doctors were selected from multiple hospitals to collect prospective colonoscopy screening videos, and the percentage of colonoscopy withdrawal overspeed was calculated to explore the relationship between it based on optical flow method and the adenoma detection rate.

Detailed description

Study showed that adequate withdrawal time is an important prerequisite for full mucosal inspection. In a large population-based analysis, a 1-min increase in withdrawal time led to a 3.6% increase in the ADR. Protocols or expert consensus recommend a standard withdrawal time of 6 min or longer. However, Findings of studies showed that a number of colonoscopies had a withdrawal time less than 6 min, which greatly reduces the ADR. Investigator's preliminary experiments have shown that deep learning can monitor the colonoscopy withdrawal time in real-time and improve the adenoma detection rate. Based on the above rich foundation of preliminary work and the massive demand for improving the colonoscopy withdrawal assessment system. The investigators improved EndoAngel to use optical flow method to monitor the colonoscopy withdrawal speed. The performance of the EndoAngel system was verified in colonoscopy videos. The investigators then aimed to evaluate whether the EndoAngel system could improve polyp detection rate after restricting the colonoscopy withdrawal speed.

Conditions

• Colonoscopy
• Artificial Intelligence
• Gastrointestinal Disease
• Deep Learning

Timeline

• Start date: 2022-07-29
• Primary completion: 2023-10-29
• Completion: 2023-11-29
• First posted: 2022-07-05
• Last updated: 2023-03-29

Locations

1 site across 1 country: China