Dear members and friends of the ESC WG on e-Cardiology,

Welcome to our August editorial, highlighting four forward-thinking contributions that illustrate how artificial intelligence (AI) and digital technologies are shaping the future of cardiovascular medicine. This month’s selection spans AI-enabled imaging, molecular profiling, ECG-based screening, and extended reality in interventional cardiology, showcasing the rich diversity of innovation in our field.

We begin with an impressive study by Miller et al. (Eur Heart J, 2025) that harnesses deep learning to quantify body composition from routine PET/CT scans [1]. In over 10,000 patients undergoing myocardial perfusion imaging, the authors used AI to extract volumetric and density metrics of skeletal muscle and various adipose tissue compartments from low-dose CT images originally acquired for attenuation correction. These body composition features were independently predictive of adverse cardiovascular outcomes, even after adjustment for traditional risk markers. In her accompanying commentary, our fellow member, Maria Chiara Gatto, highlights the clinical value of transforming standard imaging into a powerful cardiometabolic risk stratification tool. This pragmatic approach opens the door to automated reports of metabolic health that can be seamlessly integrated into existing workflows, offering new avenues for personalised prevention in patients with obesity and metabolic syndrome.

The second article, “Spatial transcriptional landscape of human heart failure” by Lee et al. (EHJ, 2025), demonstrates how advanced digital tools like spatial transcriptomics can decode the cellular and molecular heterogeneity of cardiomyopathies [2]. By integrating spatial gene expression with histology and clinical data, the study revealed region- and cell-specific disease signatures,     such as a pro-inflammatory endothelial subtype linked to fibrosis and gene changes tied to myocardial disarray. Crucially, this approach uncovered novel molecular drivers of heart failure progression (e.g. CRIP3, PFKFB2, TAX1BP3), underscoring the power of intelligent, spatially-aware profiling to enable precision diagnostics and pave the way for truly personalised cardiovascular therapies.

The third featured study, “Development and multinational validation of an ensemble deep learning algorithm for detecting and predicting structural heart disease using noisy single-lead electrocardiograms” (Aminorroaya et al., EHJ Digital Health, 2025), introduces ADAPT-HEART—a novel AI-driven tool that detects and predicts structural heart disease (SHD) using only a single-lead ECG signal from portable or wearable devices [3]. Trained on nearly 267,000 ECGs with paired echocardiographic data, the algorithm achieved excellent performance (AUROC >0.85) across multiple external validation cohorts, including population-based settings. Importantly, it retained predictive value for incident SHD, with screen-positive individuals showing up to a sixfold increased risk of future disease. In their accompanying editorial, Antoniades and Chan highlight the transformative potential of such technology to democratise cardiac diagnostics [4]. By enabling low-cost, scalable SHD screening beyond traditional care pathways, AI-ECG solutions like ADAPT-HEART can reach underserved populations and identify patients in the early, asymptomatic stages of disease—where timely interventions are most impactful. Their call to action is clear: as intelligent tools increasingly meet clinical-grade validation, we must focus on integrating them thoughtfully into real-world practice to maximise equity and efficacy in cardiovascular care.

Finally, Maria Kundzierewicz and colleagues (EHJ Digital Health, 2025) review the emerging role of extended reality (XR) — including virtual and augmented reality — in cardiovascular interventions [5]. Applications span from virtual procedural rehearsal to real-time anatomical overlays during catheterisation. XR also holds promise in medical education and patient communication. Despite current limitations in workflow integration and hardware, the potential to enhance precision, safety, and understanding is substantial.

Together, these four contributions capture the energy and evolution of e-Cardiology. We invite you to delve into each article and reflect on how these developments might influence your clinical or research practice.

On behalf of the ESC WG on e-Cardiology,
Panteleimon Pantelidis, MD
Nucleus Member, ESC WG on e-Cardiology (2024–2026)

References

1. Miller RJH, Yi J, Shanbhag A, Marcinkiewicz A, Patel KK, Lemley M, Ramirez G, Geers J, Chareonthaitawee P, Wopperer S, Berman DS, Di Carli M, Dey D, Slomka PJ. Deep learning-quantified body composition from positron emission tomography/computed tomography and cardiovascular outcomes: a multicentre study. Eur Heart J. 2025;46(24):2336-2347. 
2. Lee SE, Joo JH, Hwang HS, Chen SF, Evans D, Lee KY, Kim KH, Hyun J, Kim MS, Jung SH, Kim JJ, Lee JS, Torkamani A. Spatial transcriptional landscape of human heart failure. Eur Heart J. 2025:ehaf272. 
3. Aminorroaya A, Dhingra LS, Pedroso Camargos A, Vasisht Shankar S, Coppi A, Khunte A, Foppa M, Brant LC, Barreto SM, Ribeiro ALP, Krumholz HM, Oikonomou EK, Khera R. Development and Multinational Validation of an Ensemble Deep Learning Algorithm for Detecting and Predicting Structural Heart Disease Using Noisy Single-lead Electrocardiograms. Eur Heart J Digit Health. 2025. doi.org/10.1093/ehjdh/ztaf034 (In press)
4. Antoniades C, Chan K. Scalable screening for structural heart disease: promises from artificial intelligence-electrocardiogram tools. Eur Heart J Digit Health. 2025. doi.org/10.1093/ehjdh/ztaf048 (In press)
5. Kundzierewicz M, Kołodziej K, Khokhar A, Tsung-Ying T, Leśniak A, Zakrzewski P, Borecki H, Bohn E, Hecko J, Januska J, Precek D, Stanuch M, Skalski A, Onuma Y, Serruys P, Bruining N, Złahoda-Huzior A, Dudek D. Catheterization Laboratories open the doors for Extended Realities – review of clinical applications in cardiology. Eur Heart J Digit Health. 2025. doi.org/10.1093/ehjdh/ztaf072 (In press)