AI Algorithm Enhances Detection of High-Risk Heart Patients, Streamlining Diagnosis and Treatment
Researchers at Mount Sinai have fine-tuned an artificial intelligence (AI) algorithm to better identify individuals at high risk for hypertrophic cardiomyopathy (HCM)—a serious heart condition that can lead to sudden cardiac death. The tool, known as Viz HCM, has been adjusted to assign specific probability scores to patients flagged during electrocardiograms (ECGs), streamlining how clinicians prioritize care during appointments.
While the Viz HCM algorithm had previously gained FDA approval for spotting HCM on ECGs, the Mount Sinai team has now improved its ability to express individualized risk in numeric terms. The study, published April 22 in NEJM AI, introduces a significant enhancement: rather than just labeling a patient as “high risk” or “suspected HCM,” the algorithm can now say something like, “You have about a 60 percent chance of having HCM,” according to Dr. Joshua Lampert, Director of Machine Learning at Mount Sinai Fuster Heart Hospital.
This added granularity empowers patients with a clearer view of their personal health risk, potentially leading to faster diagnosis and targeted treatment that may prevent life-threatening complications—particularly in younger individuals.
“This is an important step forward in translating novel deep-learning algorithms into clinical practice by providing clinicians and patients with more meaningful information," said Dr. Lampert, who also serves as Assistant Professor of Medicine in Cardiology and Data-Driven and Digital Medicine at the Icahn School of Medicine at Mount Sinai.
He explained how clinicians can now use this AI-generated information to prioritize high-risk patients through clinical sorting tools, while offering patients more actionable insight via calibrated risk scores. “This can transform clinical practice because the approach provides meaningful information in a clinically pragmatic fashion to facilitate patient care,” he added.
HCM affects around 1 in 200 people worldwide, making it a key cause of heart transplants. Unfortunately, many individuals remain unaware of their condition until symptoms appear—often when the disease has already progressed.
To evaluate the algorithm, the Mount Sinai team ran Viz HCM on nearly 71,000 ECGs performed between March 2023 and January 2024. The tool flagged 1,522 patients as having a positive alert for HCM. Researchers then cross-referenced electronic health records and imaging results to confirm diagnoses.
Following confirmation, the team applied a model calibration process to verify if the predicted probabilities aligned with actual HCM diagnoses. The results showed strong alignment—demonstrating the algorithm’s ability to accurately estimate individual risk.
This calibrated model allows cardiologists to prioritize urgent cases more effectively—bringing in those most at risk for quicker evaluation and potential early treatment. Rather than vague warnings, physicians can now offer precise, personalized risk information. This level of detail may also encourage new patients to pursue follow-up care that could prevent worsening symptoms or fatal outcomes.
“This study provides much-needed granularity to help rethink how we triage, risk-stratify, and counsel patients,” said Dr. Vivek Reddy, co-senior author and Director of Cardiac Arrhythmia Services for the Mount Sinai Health System. “Using hypertrophic cardiomyopathy as an illustrative use case, we show how we can pragmatically operationalize novel tools even in the setting of less common diseases by sorting AI classifications to triage patients.”
“This study reflects pragmatic implementation science at its best, demonstrating how we can responsibly and thoughtfully integrate advanced AI tools into real-world clinical workflows,” added Dr. Girish N. Nadkarni, co-senior author and Chair of the Windreich Department of Artificial Intelligence and Human Health. “It’s not just about building a high-performing algorithm—it’s about making sure it supports clinical decision-making in a way that improves patient outcomes.”
Dr. Nadkarni emphasized that a calibrated model such as this helps clinicians identify the right patients at the right time—fulfilling AI’s potential to enhance healthcare.
The team’s next step is to expand the use of this calibrated HCM detection tool to additional healthcare systems across the U.S.
The study was sponsored by Viz.ai, and Dr. Lampert disclosed that he serves as a paid consultant for the company.
Mount Sinai’s National Leadership in Heart Care
Mount Sinai Fuster Heart Hospital currently ranks No. 4 in the U.S. for cardiology and heart surgery, as reported by U.S. News & World Report®. It also holds the top spot in New York and is ranked No. 6 globally by Newsweek.
As part of the Mount Sinai Health System—the largest academic health system in New York City—it encompasses seven hospitals, a top-tier medical school, and a network of ambulatory practices across the region. With approximately 9,000 physicians, 11 joint-venture centers, and 45 multidisciplinary institutes, Mount Sinai is recognized for its excellence in research, education, and patient-centered care. The system’s hospitals consistently appear on Newsweek’s and U.S. News & World Report’s prestigious hospital rankings.
More information is available at mountsinai.org.