EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning to predict left ventricular hypertrophy from the electrocardiogram

Hafiz Naderi iD, Thomas Kaplan, Stefan van Duijvenboden iD, Esmeralda Ruiz Pujadas iD, Nay Aung iD, C Anwar A Chahal iD, Karim Lekadir iD, Bishwas Chamling iD, Marcus Dörr iD, Marcello R P Markus iD, Steffen E Petersen iD, Julia Ramírez iD und 1 weitere

Europace · 2026

DOI: 10.1093/europace/euag015 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Europace, 2026. Verfasst von Hafiz Naderi und 12 weiteren Autor:innen. Beteiligte Einrichtung: William Harvey Research Institute, Queen Mary University of London, Charterhouse Square, London, EC1M 6BQ, UK. Hauptthemen laut MEDLINE: Hypertrophy, Left Ventricular, Electrocardiography, Deep Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: British Heart Foundation, National Institute for Health and Care Research, Integrated Academic Training Programme.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

Left ventricular hypertrophy (LVH) is a strong predictor of cardiovascular disease. We previously compared supervised machine learning techniques to classify cardiac magnetic resonance (CMR)-derived LVH using electrocardiogram (ECG) and clinical variables in 37 534 UK Biobank participants, obtaining an area under the receiving operating curve (AUROC) of 0.85, but with limited specificity and requiring external validation. In this study, we develop a deep learning (DL) model to improve classification with external evaluation in the Study of Health in Pomerania (SHIP). Methods and results We analysed 12-lead ECGs of 48 835 participants from the UK Biobank imaging study. The dataset was split into a training set (70%), validation set (15%), and test set (15%) for performance evaluation. The model architecture was a fully convolutional network, for which the input was the participants’ median ECG and clinical variables and the predicted indexed left ventricular mass (iLVM) as the output. A subsequent logistic regression model was used to recalibrate iLVM predictions. In UK Biobank, 717 (1.5%) participants had CMR-derived LVH and the AUROC for the DL model was 0.97. The ECG components most predictive of LVH were the QRS complex and ventricular rate. The DL model outperformed our supervised algorithms, previous DL modelling efforts and clinical ECG benchmarks. There was modest generalizability of the DL model to 1423 participants in SHIP (AUROC 0.78), with differences in clinical profile, ECG acquisition, and CMR labelling as important factors.

Conclusion

Our findings support the feasibility of scalable DL-based screening tools for the prediction of LVH from the ECG, whilst highlighting the need for model development using larger datasets with greater diversity to ensure generalizability.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Hypertrophy, Left VentricularElectrocardiographyDeep LearningHumansFemalePredictive Value of TestsPredictive Learning ModelsMaleMiddle AgedAgedReproducibility of ResultsClassification AlgorithmsUK BiobankConvolutional Neural NetworksPrediction AlgorithmsUnited Kingdom

Stichwörter der Autor:innen

Deep learningElectrocardiogramLeft ventricular hypertrophyMachine learning

Autor:innen und Institutionen

  • William Harvey Research Institute, Queen Mary University of London, Charterhouse Square, London, EC1M 6BQ, UK.
  • Barts Heart Centre, St Bartholomew's Hospital, Barts Health NHS Trust, West Smithfield, London, UK.
  • Big Data Institute, La Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, UK.
  • Faculty of Mathematics and Computer Science, University of Barcelona, Barcelona, Spain.
  • Centre for Inherited Cardiovascular Diseases, WellSpan Health, Lancaster, PA, USA.
  • Department of Cardiovascular Diseases, Mayo Clinic, Rochester, MN, USA.

ORCID-Profile: Hafiz Naderi, Stefan van Duijvenboden, Esmeralda Ruiz Pujadas, Nay Aung, C Anwar A Chahal, Karim Lekadir, Bishwas Chamling, Marcus Dörr, Marcello R P Markus, Steffen E Petersen, Julia Ramírez, Patricia B Munroe

Förderung

  • British Heart Foundation
  • National Institute for Health and Care Research
  • Integrated Academic Training Programme
  • European Union
  • Medical Research Council
  • Biomedical Research Centre
  • Queen Mary University of London
  • St George's University Hospitals NHS Foundation Trust

Zitieren

APA

Naderi, H., Kaplan, T., van Duijvenboden, S., Pujadas, E. R., Aung, N., Anwar A Chahal, C., Lekadir, K., Chamling, B., Dörr, M., Markus, M. R. P., Petersen, S. E., Ramírez, J., & Munroe, P. B. (2026). Deep learning to predict left ventricular hypertrophy from the electrocardiogram. Europace. https://doi.org/10.1093/europace/euag015

Vancouver

Naderi H, Kaplan T, van Duijvenboden S, Pujadas ER, Aung N, Anwar A Chahal C, et al. Deep learning to predict left ventricular hypertrophy from the electrocardiogram. Europace. 2026. doi:10.1093/europace/euag015

BibTeX
@article{Naderi2026,
  author = {Naderi, Hafiz and Kaplan, Thomas and van Duijvenboden, Stefan and Pujadas, Esmeralda Ruiz and Aung, Nay and Anwar A Chahal, C and Lekadir, Karim and Chamling, Bishwas and Dörr, Marcus and Markus, Marcello R P and Petersen, Steffen E and Ramírez, Julia and Munroe, Patricia B},
  title = {Deep learning to predict left ventricular hypertrophy from the electrocardiogram},
  journal = {Europace},
  year = {2026},
  doi = {10.1093/europace/euag015}
}
Kostenlos starten →