Dear members of the ESC WG on e-Cardiology,

We are excited to present a collection of pioneering articles demonstrating the remarkable potential of artificial intelligence (AI) and digital technologies to transform cardiovascular care, expand our understanding of heart health, and strengthen clinical practice.

Our lead article, “Artificial intelligence to improve cardiovascular population health” by Meder et al. commented in detail by Roderick W. Treskes, explores how AI can address modifiable cardiovascular risk factors and improve population health. It reviews evidence across key domains, considers implications for healthcare delivery and regulation, and offers a balanced perspective on the opportunities and challenges of integrating AI into preventive cardiology.

The second feature, “Effectiveness of Fully Immersive Virtual Reality-Based Simulation Training on Knowledge Acquisition in STEMI Emergency Management” by Einloft et al., evaluates virtual reality (VR)-based training for ST-elevation myocardial infarction (STEMI). In a single-center trial, 247 medical students were assigned to human, integrated, or no guidance while managing a virtual STEMI patient. Knowledge and clinical quality improved across all groups; human guidance outperformed no guidance but offered no advantage over integrated guidance. The findings show that VR training is feasible, effective, and that integrated guidance can serve as a practical alternative to tutor-led instruction for independent learning.

The third piece, “Role of Structural Versus Cellular Remodeling in Atrial Arrhythmogenesis: Insights From Personalized Digital Twins” by Pikunov et al., investigates how cellular remodeling drives atrial fibrillation (AF) under fibrotic conditions and develops new methods to predict reentrant driver locations. Using 3D atrial digital twins combined with pathology-specific single-cell models, the authors created a novel algorithm over 700× faster than traditional approaches. AF inducibility was similar across models, but pathological atria exhibited more arrhythmogenic substrates, with wavebreak probability increasing with fibrosis density and reentrant drivers clustering in regions of highest risk.

Finally, “AI-Enabled Sinus ECGs for Detecting Paroxysmal Atrial Fibrillation Benchmarked Against the CHARGE-AF Score” by Tarabanis et al., develops and validates a convolutional neural network (CNN) using sinus ECGs and CHARGE-AF features to predict incident paroxysmal AF. In 157,192 ECGs from 76,986 patients, the model outperformed the CHARGE-AF score (AUC 0.89, AUPRC 0.69) and maintained strong performance in US and European external cohorts and across age, sex, and race subgroups. A CNN using ECGs alone also performed well, highlighting its potential for broad AF screening.

These forward-thinking studies embody the innovative spirit guiding our board as we advance e-cardiology, foster the integration of digital technologies, and work toward improving outcomes in cardiovascular care.

On behalf of the ESC WG on e-Cardiology nucleus,

Monika Gawałko, MD, PhD,
Communication Coordinator, ESC WG on e-Cardiology 2024-2026