EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning optimisation for cardiology: Neural Architecture Search-driven arrhythmia classification with electrocardiograms

Erik Vanegas Müller iD, Arese Joe-Oshodi, Abhirup Banerjee, Mauricio Villarroel

PLOS One · 06.11.2026 · 21 (11) · e0362129

DOI: 10.1371/journal.pone.0362129 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 06.11.2026. Verfasst von Erik Vanegas Müller und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Engineering Science, The Podium Institute for Sports Medicine and Technology, University of Oxford, Oxford, United Kingdom. Hauptthemen laut MEDLINE: Deep Learning, Arrhythmias, Cardiac, Electrocardiography, Cardiology. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Podium Institute of Sports Medicine and Technology, Royal Society University Research Fellowship.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cardiovascular disease is the leading cause of death worldwide. Sudden cardiac death (SCD) accounts for roughly 50% of all cardiac deaths. The electrocardiogram (ECG) is widely used for diagnosing cardiac disease. However, the complexity of interpretation limits the ECG’s efficacy. Modern deep learning methods can assist clinicians in diagnosis. We applied Neural Architecture Search (NAS) to identify optimal deep learning architectures for classifying cardiac arrhythmias with ECGs. We applied the Differentiable Architecture Search strategy to an AutoFormer search and trained, validated, and tested the resulting model on the PhysioNet Challenge 2021 (PC21) dataset (n = 88,253 from three continents). We optimised hyperparameters on the NAS output, exploring input patch size, class weighting, and loss function. The PC21 metric and the area under the receiver operating characteristic curve (AUROC) are the performance metrics. The NAS converged towards minimal architectural configurations (embedding dimension: 384, depth: 4, self-attention heads: 4, MLP ratio: 1) with a validation and test PC21 metric of 0.66 (PC21 winner: 0.63). The NAS-created network achieved an AUROC of 0.97 and a PC21 metric of 0.72 during testing (PC21 winner: 0.69). We tested the PC21 winner architecture (1D-convolutional neural network with attention mechanism) without the attention mechanism and with the NAS-suggested architecture of 4 self-attention heads (instead of 8). The architecture without the attention mechanism performed similarly to the one with 8 self-attention heads (AUROC: 0.90, PC21 metric: 0.69). The architecture with 4 self-attention heads scored an AUROC of 0.91 and improved the PC21 metric to 0.71. The attention mechanism visualisation for two SCD-related arrhythmias (premature ventricular contractions and prolonged QT-interval) showed more clinically relevant attention for the 4-self-attention-head architectures. We interpret architectural simplicity as driving performance in arrhythmia classification. The self-attention mechanism can autonomously abstract clinical representations, simplifying clinical deployment by eliminating the need for an explicit feature-extraction pipeline.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningArrhythmias, CardiacElectrocardiographyCardiologyHumansNeural Networks, ComputerROC Curve

Autor:innen und Institutionen

  • Department of Engineering Science, The Podium Institute for Sports Medicine and Technology, University of Oxford, Oxford, United Kingdom.
  • Department of Engineering Science, Institute of Biomedical Engineering, University of Oxford, Oxford, United Kingdom.

ORCID-Profile: Erik Vanegas Müller

Förderung

  • The Podium Institute of Sports Medicine and Technology
  • Royal Society University Research Fellowship

Zitieren

APA

Vanegas Müller, E., Joe-Oshodi, A., Banerjee, A., & Villarroel, M. (2026). Deep learning optimisation for cardiology: Neural Architecture Search-driven arrhythmia classification with electrocardiograms. PLOS One, 21(11), e0362129. https://doi.org/10.1371/journal.pone.0362129

Vancouver

Vanegas Müller E, Joe-Oshodi A, Banerjee A, Villarroel M. Deep learning optimisation for cardiology: Neural Architecture Search-driven arrhythmia classification with electrocardiograms. PLOS One. 2026;21(11):e0362129. doi:10.1371/journal.pone.0362129

BibTeX
@article{VanegasMller2026,
  author = {Vanegas Müller, Erik and Joe-Oshodi, Arese and Banerjee, Abhirup and Villarroel, Mauricio},
  title = {Deep learning optimisation for cardiology: Neural Architecture Search-driven arrhythmia classification with electrocardiograms},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {11},
  pages = {e0362129},
  doi = {10.1371/journal.pone.0362129}
}
Kostenlos starten →