Actively Recruiting
Development and Clinical Use of AI-Powered Electrocardiography for Detecting and Predicting Cardiovascular Diseases A Multicenter Retrospective Study
Led by Inha University Hospital · Updated on 2026-10-02
10000
Participants Needed
1
Research Sites
N/A
Total Duration
AI-Summary
What This Trial Is About
Researchers are investigating how an artificial intelligence AI system can help diagnose and predict outcomes for people with cardiovascular diseases such as coronary artery disease and heart failure. This study is retrospective and multicenter, using data collected from multiple tertiary medical centers in South Korea. It focuses on improving the use of standard 12-lead electrocardiograms ECGs by applying deep learning technologies to enhance detection of heart conditions and forecast clinical results. The study analyzes encrypted raw 12-lead ECG data in text XML format using advanced deep neural network models that combine Bi-LSTM and attention mechanisms. These AI models are developed and validated to diagnose conditions like coronary artery disease, heart failure, atrial fibrillation, and ventricular hypertrophy. The research uses data from patient records spanning 2005 to 2022 to assess how well AI-enhanced ECGs perform in clinical settings. Participants are patients aged 18 or older with cardiovascular diseases who have available 12-lead ECG records. The study involves extracting and encrypting ECG data to ensure privacy, then applying AI models to analyze and predict disease presence and progression. Researchers measure the diagnostic accuracy of AI-ECG models using metrics like AUC-ROC, sensitivity, and specificity. The study runs from March 2023 until March 2028, with no active treatment involved as it is observational and retrospective.
CONDITIONS
Brief Title
AI-Powered Electrocardiography for Diagnosis and Prognostic Prediction in Cardiovascular Disease (AI-CVD): Multicenter Retrospective Study
Research Team
Y
Yongsoo Baek, MD, PhD
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here