Evaluating AI Software Accuracy for Heart Pumping Function Using the Peerbridge COR ECG Wearable Monitor

What this Trial Is About

Researchers are evaluating an investigational artificial intelligence AI software designed to estimate the severity of ejection fraction EF, which indicates how well the heart pumps blood. This prospective, multicenter, cluster-randomized controlled study compares EF severity categories determined by the AI software using continuous ECG waveform data to those measured by an FDA-cleared transthoracic echocardiogram TTE. The study aims to provide a low-burden, cost-effective alternative for EF monitoring in heart failure and related heart conditions, especially where traditional imaging access is limited. Participants will use the FDA-cleared Peerbridge COR4 ECG Wearable Monitor, a patch device worn during daily activities, to collect ECG data. During a 15-minute resting session while seated upright, 5-minute ECG segments will be recorded and analyzed by the AI software to estimate EF severity based on the American Society of Echocardiographys scale. The study includes two subprotocols one with 30 minutes of ECG recording including 15 minutes analyzed, and another allowing up to 7 days of device use with periodic sitting sessions. The EF severity from the AI software will be compared against results from echocardiography. Participants will be enrolled at multiple sites, providing paired data points consisting of simultaneous or near-simultaneous ECG recordings and echocardiograms. They will follow a standardized 15-minute seated session protocol using the wearable device, pressing an event button to mark the session start and end. Data collection includes medical histories, 12-lead ECGs, and device logs. The main outcome measures focus on agreement between the AI softwares EF severity categories and those from echocardiography over an average of 9 months. Safety and compliance will be monitored throughout the study period.

AI-Enabled Direct-from-ECG Ejection Fraction (EF) Severity Assessment Using COR ECG Wearable Monitor