mlsanityData to 7 Oct 2026

Deep Learning for ECG Analysis: Benchmarks and Insights from PTB-XL

arXiv 2004.13701 ↗First posted 28 Apr 2020cs.LG stat.MLcited by 158

Not on the map. Why?

Abstract (arXiv)

Electrocardiography is a very common, non-invasive diagnostic procedure and its interpretation is increasingly supported by automatic interpretation algorithms. The progress in the field of automatic ECG interpretation has up to now been hampered by a lack of appropriate datasets for training as well as a lack of well-defined evaluation procedures to ensure comparability of different algorithms. To alleviate these issues, we put forward first benchmarking results for the recently published, freely accessible PTB-XL dataset, covering a variety of tasks from different ECG statement prediction tasks over age and gender prediction to signal quality assessment. We find that convolutional neural networks, in particular resnet- and inception-based architectures, show the strongest performance across all tasks outperforming feature-based algorithms by a large margin. These results are complemented by deeper insights into the classification algorithm in terms of hidden stratification, model uncertainty and an exploratory interpretability analysis. We also put forward benchmarking results for the ICBEB2018 challenge ECG dataset and discuss prospects of transfer learning using classifiers pretrained on PTB-XL. With this resource, we aim to establish the PTB-XL dataset as a resource for structured benchmarking of ECG analysis algorithms and encourage other researchers in the field to join these efforts.

Key facts. Cited by 158 papers in the map (data through 7 Oct 2026). First posted on arXiv 28 Apr 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 10; 2023: 32; 2024: 24; 2025: 36; 2026: 51.

158 papers.

Sort:

By area:

  1. 2023Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlookcited by 109
  2. 2024Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogramcited by 84
  3. 2021Self-supervised representation learning from 12-lead ECG datacited by 80
  4. 2023Effectively Modeling Time Series with Simple Discrete State Spacescited by 54
  5. 2023FedMultimodal: A Benchmark For Multimodal Federated Learningcited by 51
  6. 2025Reading Your Heart: Learning ECG Words and Sentences via Pre-training ECG Language Modelcited by 36
  7. 2023Diffusion-based Conditional ECG Generation with Structured State Space Modelscited by 35
  8. 2024Variational Diffusion Posterior Sampling with Midpoint Guidancecited by 32
  9. 2022Modeling Multivariate Biosignals With Graph Neural Networks and Structured State Space Modelscited by 29
  10. 2024An Electrocardiogram Foundation Model Built on over 10 Million Recordings with External Evaluation across Multiple Domainscited by 27
  11. 2024ECG Semantic Integrator (ESI): A Foundation ECG Model Pretrained with LLM-Enhanced Cardiological Textcited by 25
  12. 2023Explainable AI for Time Series via Virtual Inspection Layerscited by 23
38 more
  1. 2023Transfer Knowledge from Natural Language to Electrocardiography: Can We Detect Cardiovascular Disease Through Language Models?cited by 21
  2. 2023Sequence Modeling with Multiresolution Convolutional Memorycited by 19
  3. 2023Explaining Deep Learning for ECG Analysis: Building Blocks for Auditing and Knowledge Discoverycited by 15
  4. 2023Towards quantitative precision for ECG analysis: Leveraging state space models, self-supervision and patient metadatacited by 15
  5. 2024Self-Supervised Pre-Training with Joint-Embedding Predictive Architecture Boosts ECG Classification Performancecited by 14
  6. 2023Towards Enhancing Time Series Contrastive Learning: A Dynamic Bad Pair Mining Approachcited by 13
  7. 2023ECGBERT: Understanding Hidden Language of ECGs with Self-Supervised Representation Learningcited by 10
  8. 2024Communication-Efficient Multimodal Federated Learning: Joint Modality and Client Selectioncited by 10
  9. 2024Transfer Learning for Diffusion Modelscited by 10
  10. 2022Advancing the State-of-the-Art for ECG Analysis through Structured State Space Modelscited by 9
  11. 2022PulseImpute: A Novel Benchmark Task for Pulsative Physiological Signal Imputationcited by 9
  12. 2024Multi-Channel Masked Autoencoder and Comprehensive Evaluations for Reconstructing 12-Lead ECG from Arbitrary Single-Lead ECGcited by 9
  13. 2024ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modelingcited by 9
  14. 2025Generalizable deep learning for photoplethysmography-based blood pressure estimation -- A Benchmarking Studycited by 9
  15. 2022GeoECG: Data Augmentation via Wasserstein Geodesic Perturbation for Robust Electrocardiogram Predictioncited by 8
  16. 2024Large Language Model-informed ECG Dual Attention Network for Heart Failure Risk Predictioncited by 8
  17. 2025Machine-learning for photoplethysmography analysis: Benchmarking feature, image, and signal-based approachescited by 8
  18. 2021Robustness of convolutional neural networks to physiological ECG noisecited by 7
  19. 2022On the role of benchmarking data sets and simulations in method comparison studiescited by 7
  20. 2023In-Distribution and Out-of-Distribution Self-supervised ECG Representation Learning for Arrhythmia Detectioncited by 7
  21. 2023Masked Transformer for Electrocardiogram Classificationcited by 7
  22. 2023Scaling Representation Learning from Ubiquitous ECG with State-Space Modelscited by 7
  23. 2023Using explainable AI to investigate electrocardiogram changes during healthy aging -- from expert features to raw signalscited by 7
  24. 2023Uncertainty Quantification in Machine Learning for Biosignal Applications -- A Reviewcited by 7
  25. 2022MedalCare-XL: 16,900 healthy and pathological 12 lead ECGs obtained through electrophysiological simulationscited by 6
  26. 2023Semi-Supervised Learning for Multi-Label Cardiovascular Diseases Prediction:A Multi-Dataset Studycited by 6
  27. 2025SuPreME: A Supervised Pre-training Framework for Multimodal ECG Representation Learningcited by 6
  28. 2025Explainable Deep Neural Network for Multimodal ECG Signals: Intermediate vs Late Fusioncited by 6
  29. 2025Benchmarking ECG FMs: A Reality Check Across Clinical Taskscited by 6
  30. 2023MVKT-ECG: Efficient Single-lead ECG Classification on Multi-Label Arrhythmia by Multi-View Knowledge Transferringcited by 5
  31. 2024Estimation of Cardiac and Non-cardiac Diagnosis from Electrocardiogram Featurescited by 5
  32. 2024AnyECG: Foundational Models for Multitask Cardiac Analysis in Real-World Settingscited by 5
  33. 2023Spectral Cross-Domain Neural Network with Soft-adaptive Threshold Spectral Enhancementcited by 4
  34. 2023Deep Neural Networks Generalization and Fine-Tuning for 12-lead ECG Classificationcited by 4
  35. 2023Transforming ECG Diagnosis:An In-depth Review of Transformer-based DeepLearning Models in Cardiovascular Disease Detectioncited by 4
  36. 2023IncepSE: Leveraging InceptionTime's performance with Squeeze and Excitation mechanism in ECG analysiscited by 4
  37. 2025A metrological framework for uncertainty evaluation in machine learning classification modelscited by 4
  38. 2025ProtoECGNet: Case-Based Interpretable Deep Learning for Multi-Label ECG Classification with Contrastive Learningcited by 4

50 most cited of 158 All 158 (JSON)