thank you. Please listen to this article using the above players. ✖
Want to listen to this article for free?
Complete the form below to unlock access to all audio articles.
New AI (AI) models predict which patients are at higher risk of sudden cardiac arrest, significantly improving that they outweigh the accuracy of traditional clinical methods. The federally funded research led by Johns Hopkins University, focuses on hypertrophic cardiomyopathy, a genetic cardiac condition that can lead to sudden cardiac death, particularly in young people and athletes.
Enhanced risk prediction through medical imaging
A model of multimodal AI for ventricular arrhythmias risk stratification (MAAR) analyzes a variety of medical records and cardiac imaging to assess the risk of sudden cardiac death in patients. Although current clinical guidelines used by physicians only have a 50% accuracy rate in identifying high-risk patients, MAARS exceeds these guidelines, exceeding 89% in all patients and up to 93% in patients aged 40-60.
The breakthrough comes from the model's ability to examine MRI images of the heart with enhanced contrast. This is a technology that was not previously used in such details. AI identifies cardiac scaling-fibrosis patterns that are at high risk but often overlooked in traditional assessments. This ability is important as it is a key marker of sudden cardiomyopathy patients, but it has been difficult for physicians to interpret live MRI images.
Possible life-saving intervention
This AI-driven model helps physicians better target interventions. By predicting which patients are at the greatest risk, we can potentially save lives by recommending precautions such as defibrillator implantation. On the other hand, it can also prevent unnecessary treatment for patients who do not require such intervention.
The research team's model offers the additional advantage of explaining why certain patients are considered high risk and why physicians can develop personalized treatment plans based on their individual needs. This allows for changes in clinical care by improving the accuracy and effectiveness of medical interventions.
Expand the model's functionality
The Johns Hopkins team plans to continue testing AI models on additional patient groups and hopes to extend its applications to other cardiac diseases, such as cardiac sarcoidosis and arrhythmic right ventricular cardiomyopathy. Their previous studies have also demonstrated the possibility of AI to evaluate survival predictions in patients with cardiac infarction, further demonstrating the promise of these models in cardiac disease.
The authors of this study include Chang Singh Rai, Mingalan Ying, Eugene G. Holmovski, Dan M. Popeski, Edem Binka, Stephen L. Zimmerman, and Erica Scherer and Dermot M. Phelan of Alison G. Atrium Health, Johns Hopkins. This study was supported by National Institutes of Health grants and LEDUCQ Foundation grants.
reference: Rogers AJ, Reynbakh O, Ahmed A, et al. Cardiovascular imaging technology for electrophysiologists. Nat Cardiovasc Res. 2025. doi:10.1038/s44161-025-00648-8
This article has been republished from: Note: Materials may have been edited for length and content. For more information, please contact the source quoted. The press release publishing policy can be accessed here.
This content contains text generated with AI support. The AI policy of the Technology Network can be found here.
