NCT06617468

Effect of the Computer Aided Diagnosis with Explainable Artificial Intelligence for Colon Polyp on Optical Diagnosis and Acceptance of Technology

• Status: Active Not Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 120 (estimated)
• Sponsor: Seoul National University Hospital · Academic / Other
• Sex: All
• Age: —
• Healthy volunteers: Not accepted

Summary

The goal of this clinical trial is to learn if computer-aided diagnosis with deep learning and computer-aided diagnosis with explainable AI work to optical diagnosis performance and acceptance of technology in endoscopists. The main questions it aims to answer are: Do computer-aided diagnosis with deep learning and computer-aided diagnosis with explainable AI improve optical diagnosis performance in endoscopists? Does experience using deep learning-based computer-assisted diagnosis and explainable AI-based computer-assisted diagnosis improve endoscopists' acceptance of computer-aided diagnosis as a technology? Participants will: Conduct a survey on acceptance and use of technology about computer-aided diagnosis. Perform a test to estimate the pathologic diagnosis on 200 NBI still images without the aid of computer-aided diagnosis. More than 1 month later, perform a same test to estimate the pathologic diagnosis on 200 NBI still images with computer-aided diagnosis with deep learning or explainable AI. Conduct a survey on acceptance and use of technology about computer-aided diagnosis.

Conditions

• Colon Polyp

Interventions

Type	Name	Description
DIAGNOSTIC_TEST	computer-aided diagnosis with explainable AI	Perform a same test to estimate the pathologic diagnosis on 200 NBI still images with computer-aided diagnosis with explainable AI.
DIAGNOSTIC_TEST	computer-aided diagnosis with deep learning	Perform a same test to estimate the pathologic diagnosis on 200 NBI still images with computer-aided diagnosis with deep Iearning.

Timeline

• Start date: 2024-09-20
• Primary completion: 2024-12-31
• Completion: 2024-12-31
• First posted: 2024-09-27
• Last updated: 2024-10-04

Locations

1 site across 1 country: South Korea