NCT04749927

Deep Learning of Retinal Photographs and Atherosclerotic Cardiovascular Disease

Prediction of Incident Atherosclerotic Cardiovascular Disease From Retinal Photographs Via Deep Learning

• Status: Recruiting
• Phase: —
• Study type: Observational
• Enrollment: 2,400 (estimated)
• Sponsor: Yonsei University · Academic / Other
• Sex: All
• Age: 20 Years – 79 Years
• Healthy volunteers: Accepted

Summary

The research team has developed a deep learning algorithm that predicts anthropometric factors from fundus photographs and an algorithm that predicts cardiovascular disease risk. Fundus photographs are taken for various cardiovascular diseases (myocardial infarction, heart failure, hypertension with target organ damage, high-risk dyslipidemia, diabetic patients, and low-risk hypertension patients), and a deep learning algorithm for predicting developed anthropometric factors will be validated. Fundus photographs will also be taken twice in the first year, and additional fundus photographs will be taken two years later. Major cardiovascular events will be followed up for 5 years to verify the deep learning algorithm predicting cardiovascular disease risk prospectively.

Conditions

• Cardiovascular Disease

Timeline

• Start date: 2020-10-11
• Primary completion: 2029-10-10
• Completion: 2029-10-10
• First posted: 2021-02-11
• Last updated: 2021-02-11

Locations

1 site across 1 country: South Korea