Developing and Validating AI Tools for ECG Analysis to Detect Dangerous Heart Rhythms and Immune Therapy-Related Myocarditis

What this Trial Is About

Researchers are conducting an observational study called ELDORA to develop and validate artificial intelligence AI tools that analyze electrocardiogram ECG data for detecting life-threatening heart rhythm problems, such as Torsades-de-Pointes and immune checkpoint inhibitor-induced myocarditis. This project focuses on standardizing diverse ECG records from both analog and digital sources and aims to support clinical research in predicting risks related to these conditions. The study consolidates ECG data and clinical information from about 49 national and international cohorts, gathering roughly 127,000 subjects and up to 10 million ECG recordings. These datasets include a wide range of health conditions and populations, including healthy volunteers, patients with long QT syndrome or cancer treated with immune therapies, and those with cardiovascular or metabolic disorders. The ECG data will be harmonized into a unified database called ECGInsight, and AI models will be trained and evaluated using this comprehensive resource. Participants are involved through their existing ECG data and clinical records no treatments or interventions are assigned during the study. Researchers will assess AI model performance using various metrics like accuracy, sensitivity, specificity, and others throughout the study period, which is anticipated to last up to 48 months. The study complies with data protection regulations and uses de-identified information, with no direct impact on patient care during the research.

AI-powered ECG Analysis for Deadly Arrhythmias and ICI Myocarditis