AI ECG Algorithm for Detecting LV Systolic Dysfunction

This prospective observational cohort study aims to evaluate the clinical performance of a deep learning-based electrocardiography (ECG) algorithm (DeepECG LVSD) for detecting left ventricular systolic dysfunction (LVSD), defined as left ventricular ejection fraction (LVEF) ≤40%, using transthoracic echocardiography as the reference standard. Approximately 15,000 adult patients undergoing both ECG and echocardiography within 30 days at Ajou University Hospital will be enrolled. Diagnostic performance will be assessed using the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Secondary analyses will evaluate the association between AI-predicted LVSD and 30-day clinical outcomes, including all-cause mortality, emergency department visits, and heart failure rehospitalization.

NCT Number	NCT07636759
Lead Sponsor	Ajou University School of Medicine Collaborators: VUNO Inc.
Conditions	HF - Heart Failure
Enrollment	1,500 participants
Start Date	2026-01-01
Primary Completion	2027-12-31 (estimated)
Study Completion	2027-12-31 (estimated)
Updated on ClinicalTrials.gov	2026-08-07