Studies Support AI-ECG for Detecting Suspected Hypertrophic Cardiomyopathy

Hypertrophic cardiomyopathy is frequently underdiagnosed despite being an important cause of cardiac morbidity. Routine 12-lead electrocardiograms are widely available in emergency, inpatient, and perioperative settings, but subtle signs of the disease may be missed with standard interpretation. Earlier identification could help accelerate specialty evaluation and risk assessment. New studies presented at the American Heart Association’s Scientific Sessions 2026 evaluate an AI-enabled electrocardiogram system designed to flag suspected hypertrophic cardiomyopathy from routine tracings.

Viz HCM, an algorithm within the Viz.ai Viz Cardio Suite, is designed to help clinicians detect signs of hypertrophic cardiomyopathy (HCM) from a standard 12-lead electrocardiogram. The tool analyzes ECG data to flag suspected disease that may warrant further evaluation. Its development was supported through a multi-year agreement between Viz.ai and Bristol Myers Squibb focused on improving the identification and management of HCM.

The U.S. Food and Drug Administration (FDA) granted De Novo clearance (DEN230003) to Viz HCM in August 2023, making it the first FDA-cleared artificial intelligence algorithm intended to help clinicians detect signs of HCM on a standard 12-lead ECG. Since then, the technology has been evaluated across academic and real-world clinical settings, with 21 clinical abstracts and studies presented to date.

Two additional research abstracts were accepted for presentation at the Hypertrophic Cardiomyopathy Medical Society meeting, held in conjunction with the American Heart Association (AHA) Scientific Sessions 2026 in Chicago. Investigators from Beth Israel Deaconess Medical Center and the University of Colorado examined whether AI-ECG could identify patients who might be missed by traditional detection methods. The studies use real-world ECG data and compare the algorithm with standard ECG measurements and diagnosis code-based approaches.

In a University of Colorado analysis of more than 29,000 patients, nearly 40% of individuals flagged by Viz HCM for suspected HCM had no corresponding diagnosis code in their medical records, suggesting that AI-ECG may complement code-based identification by surfacing additional patients for specialty evaluation. 

In a separate Beth Israel Deaconess study, the AI-ECG algorithm identified suspected HCM beyond what standard ECG measurements detected, and patients flagged by the algorithm had a 2.68-fold higher hazard of adverse clinical events. Support for the current research was provided by MyoKardia, Inc., a Bristol Myers Squibb company, through a funding agreement with Viz.ai.

“We've been using ECGs for decades, but we're still discovering just how much they can tell us. In our study, Viz HCM added meaningful information beyond standard ECG interpretation in identifying patients with suspected disease. This opens up exciting possibilities for how we use a test we already rely on every day to better evaluate and care for patients,” said Jordan B. Strom, MD, MSc, of Beth Israel Deaconess Medical Center.

“Most patients with HCM are never diagnosed, and the ones who are often wait years. What these two studies show is that the ECGs that health systems are already collecting contain the signal to change that,” said Tim Showalter, MD, Chief Medical Officer at Viz.ai. “Collaborations like these are essential to advancing our understanding of what’s possible with the technology within the Viz Cardio Suite and bringing these innovations into everyday clinical care, where they can make a meaningful difference for patients.”

Related Links
Viz.ai