AI-driven outcome modelling and multimodal data integration in cardiovascular medicine

Cardiovascular medicine is rapidly transforming through the integration of longitudinal, high-dimensional, and multimodal data. Electronic health records, genomics, molecular biomarkers, imaging, ECG signals, hemodynamic data, wearable sensors, and patient-reported outcomes together provide a comprehensive understanding of cardiovascular risk, disease progression, and treatment response. Yet, synthesizing these heterogeneous data sources into clinically meaningful, reliable, and generalizable tools for patient care remains a substantial challenge. Artificial intelligence (AI)—encompassing machine learning, deep learning, and foundation models—offers the means to move beyond traditional risk calculators and toward integrative outcome modelling frameworks that can enable precision prevention, diagnosis, monitoring, and therapeutic decision-making across the cardiovascular continuum. However, reproducibility, interpretability, and clinical readiness remain critical barriers requiring further research.This Research Topic aims to highlight novel advances in AI-driven outcome modelling and multimodal data integration in cardiovascular medicine. We welcome studies that develop, validate, or translate advanced computational frameworks capable of predicting clinical outcomes, classifying disease states, or supporting treatment optimization. Submissions may address methods that combine diverse data modalities or introduce state-of-the-art algorithms applied to a single modality, provided they contribute methodological innovation or clinical translational value. We are particularly interested in research addressing coronary artery disease, heart failure, arrhythmias, hypertension, valvular heart disease, cardio-oncology, cardiometabolic disorders, and vascular medicine, as well as studies demonstrating integration across imaging, molecular, and clinical contexts for improved decision support.To advance current understanding and facilitate translation into clinical care, this Research Topic invites original research, reviews, and methods papers that explore—but are not limited to the following themes:-AI-driven prediction of cardiovascular outcomes, including mortality, hospitalization, treatment response, complications, and disease progression-Multimodal cardiovascular data integration across EHRs, imaging, ECG/EKG, echocardiography, MRI, CT, biomarkers, genomics, wearables, and patient-reported outcomes-Novel machine learning, deep learning, foundation, survival, and causal models for cardiovascular outcome prediction and disease stratification-Approaches to explainable, interpretable, and ethically sound AI for clinical decision support-Longitudinal modelling of cardiovascular disease trajectories, treatment pathways, and dynamic risk patterns-Validation and benchmarking of models across institutions, populations, modalities, and healthcare systems-Translational and regulatory perspectives on moving AI-driven cardiovascular models from development to clinical implementation Topic editors: Amgad Muneer, Hui Xu Source: https://www.frontiersin.org/research-topics/84534/ai-driven-outcome-modelling-and-multimodal-data-integration-in-cardiovascular-medicine