New AI model outperforms humans at predicting sudden cardiac death
A new artificial intelligence model may be able to predict patients at greatest risk of cardiac arrest with greater accuracy than human providers.
The model spots previously undetected scarring patterns on prior cardiac MRI scans of patients with hypertrophic cardiomyopathy—a leading cause of sudden cardiac death among young people. Experts involved in its development believe the tool could significantly improve the ability to appropriately manage these patients, as it provides new information providers were previously unable to identify on their own.
“Currently we have patients dying in the prime of their life because they aren’t protected and others who are putting up with defibrillators for the rest of their lives with no benefit,” senior author Natalia Trayanova, PhD, a professor of medicine at the Johns Hopkins School of Medicine, noted. “We have the ability to predict with very high accuracy whether a patient is at very high risk for sudden cardiac death or not.”
Patients with hypertrophic cardiomyopathy develop scarring, or fibrosis, of the heart over time. This scarring is what makes them most susceptible to sudden cardiac death, but it can be difficult to quantify. The Multimodal AI for ventricular Arrhythmia Risk Stratification (MAARS) model combines patients’ clinical information with their contrast-enhanced MRI scans to identify and quantify scarring patterns in a way that was previously not possible.
“People have not used deep learning on those images,” the authors explained. “We are able to extract this hidden information in the images that is not usually accounted for.”
In real-time clinical testing, MAARS predictions were compared alongside the use of current clinical guidelines on a group of patients being seen at Johns Hopkins Hospital and Sanger Heart & Vascular Institute in North Carolina. While current guidelines were accurate around 50% of the time in identifying patients at greatest risk of cardiac death, MAARS achieved 89% accuracy. Its performance improved in patients between 40 and 60, accurately predicting outcomes in 93% of cases. This was especially impressive, the researchers suggested, since this age group has long been considered to have greatest risk of death related to hypertrophic cardiomyopathy.
“Our study demonstrates that the AI model significantly enhances our ability to predict those at highest risk compared to our current algorithms and thus has the power to transform clinical care,” the group noted.
The team plans to continue testing the model’s capabilities in the future and intends to expand their research into other types of heart disease.
Learn more about the study here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.