Artificial Intelligence in Cardiomyopathies: From Opportunistic Detection to Decision-Critical Risk Stratification.
Authors
Affiliations (5)
Affiliations (5)
- Department of Translational Medical Sciences, Inherited and Rare Cardiovascular Diseases, University of Campania 'Luigi Vanvitelli', Monaldi Hospital, Via Leonardo Bianchi, Naples 80131, Italy; Department of Cardiology, Sahlgrenska University Hospital, Blå stråket 5, Gothenburg SE-413 45, Sweden; Institute of Medicine at Sahlgrenska Academy, University of Gothenburg, Wallenberg Laboratory, Blå stråket 5, Gothenburg SE-413 45, Sweden.
- Department of Translational Medical Sciences, Inherited and Rare Cardiovascular Diseases, University of Campania 'Luigi Vanvitelli', Monaldi Hospital, Via Leonardo Bianchi, Naples 80131, Italy.
- Department of Engineering, Research Unit of Intelligent Technology for Health and Wellbeing, Università Campus Bio-Medico di Roma, Via Alvaro del Portillo, 21, Rome 00128, Italy.
- Department of Engireering, Intelligent Health Technology Lab, Universitario Campus Bio-Medico, Via Alvaro del Portillo, 21, Rome 00128, Italy; Fondazione Policlinico Universitario Campus Bio-Medico, Via Alvaro del Portillo, 200, Rome 00128, Italy.
- Department of Translational Medical Sciences, Inherited and Rare Cardiovascular Diseases, University of Campania 'Luigi Vanvitelli', Monaldi Hospital, Via Leonardo Bianchi, Naples 80131, Italy; Institute of Cardiovascular Science, University College London, Gower Street, London WC1E 6BT, UK. Electronic address: [email protected].
Abstract
This article examines the evolving role of artificial intelligence (AI) across the continuum of care in cardiomyopathies, from opportunistic screening to decision-critical risk stratification. AI-enhanced electrocardiography enables scalable early detection, while advanced echocardiographic and cardiac magnetic resonance-based radiomics refine myocardial substrate characterization. Multimodal models integrating imaging, electrical, and clinical data refine risk stratification beyond conventional metrics such as left ventricular ejection fraction. However, clinical translation requires external validation, calibration stability, interpretability, and structured workflow integration. AI emerges not as autonomous diagnosis, but as accountable, clinician-supervised decision support embedded within multidisciplinary cardiovascular care.