Evaluation of Electrocardiographic Data From High-risk Cardiac Patients Using Willem Cardiologist-level Artificial Intelligence Software. WILLEM Trial.

What this Trial Is About

This research investigates the Willem AI-powered ECG analysis platform designed to detect arrhythmias and abnormal cardiac patterns in high-risk cardiac patients. The study aims to evaluate whether this AI tool can classify and predict arrhythmic episodes at the level of a cardiologist and potentially delay or prevent serious cardiac events like sudden death. The study includes patients with cardiac arrhythmias or abnormal ECG patterns and requires at least one ECG record with over one year of follow-up data. The study is observational and multicenter, comparing AI-driven ECG analysis with expert cardiologist diagnoses. Patients are divided into a training group, where new algorithms are developed, and a test group, which evaluates these algorithms to avoid overfitting. The platform analyzes electrical cardiac signals lasting 10 seconds or more from various medical devices, including hospital 12-lead ECGs, telemetries, Holter monitors, wearable ECG devices, and mobile telemetry patches. Participants will provide ECG data and clinical information, which expert cardiologists will label for arrhythmias. The study tracks outcomes such as arrhythmia detection time from real-time to 7 minutes, survival, major adverse cardiovascular and cerebrovascular events, rehospitalization, and quality of life over one year after the first ECG or enrollment. Consent is required, and patients unable to consent may have a relative provide it. The study will assess the AI platforms accuracy, sensitivity, and specificity in detecting cardiac abnormalities.

AI-powered ECG Analysis Using Willem™ Software in High-risk Cardiac Patients (WILLEM)