SANTA CLARA, Calif.– HeartBeam, Inc. (NASDAQ: BEAT) said Monday that new study data presented at HRX Live 2025 in Atlanta demonstrated its deep learning algorithms achieved high accuracy in detecting atrial fibrillation, atrial flutter, and sinus rhythm, performing comparably on both the company’s HeartBeam 3D ECG system and standard 12-lead ECGs.
In the study, ECG readings were collected from 201 patients using both devices. HeartBeam’s AI, trained on more than 10,000 standard 12-lead ECGs, was then compared to diagnoses made by a panel of three electrophysiologists. Researchers found no significant differences in accuracy between the two groups, with results showing 94.5 percent accuracy for HeartBeam’s system versus 95.5 percent for standard ECGs.
“This study represents an exciting step forward in making advanced cardiac monitoring more user-friendly and widespread,” said Rob Eno, CEO of HeartBeam. “The comparable performance of our deep learning algorithms applied to a credit card-sized device and traditional 12-lead ECG systems opens new avenues for patient care, particularly in environments where full ECG setups are impractical.”
The company said the findings will help support future FDA submissions aimed at expanding its product offerings. HeartBeam’s 3D ECG technology, designed to provide cable-free monitoring from three non-coplanar directions, received FDA clearance for arrhythmia assessment in December 2024. Its 12-lead ECG synthesis software remains under FDA review.
Dr. Joshua Lampert, a cardiac electrophysiologist at Mount Sinai Heart, presented the study results, underscoring the potential for portable ECG devices to broaden access to early arrhythmia detection and monitoring.
