mlsanityData to 7 Oct 2026

Learning General Representation of 12-Lead Electrocardiogram with a Joint-Embedding Predictive Architecture

arXiv 2410.08559 ↗First posted 11 Oct 2024cs.LG cs.AIcited by 17

Not on the map. Why?

Abstract (arXiv)

Electrocardiogram (ECG) captures the heart's electrical signals, offering valuable information for diagnosing cardiac conditions. However, the scarcity of labeled data makes it challenging to fully leverage supervised learning in the medical domain. Self-supervised learning (SSL) offers a promising solution, enabling models to learn from unlabeled data and uncover meaningful patterns. In this paper, we show that masked modeling in the latent space can be a powerful alternative to existing self-supervised methods in the ECG domain. We introduce ECG-JEPA, an SSL model for 12-lead ECG analysis that learns semantic representations of ECG data by predicting in the hidden latent space, bypassing the need to reconstruct raw signals. This approach offers several advantages in the ECG domain: (1) it avoids producing unnecessary details, such as noise, which is common in ECG; and (2) it addresses the limitations of naive L2 loss between raw signals. Another key contribution is the introduction of Cross-Pattern Attention (CroPA), a specialized masked attention mechanism tailored for 12-lead ECG data. ECG-JEPA is trained on the union of several open ECG datasets, totaling approximately 180,000 samples, and achieves state-of-the-art performance in various downstream tasks including diagnostic classification, feature extraction, and segmentation. Our code is openly available at https://github.com/sehunfromdaegu/ECG_JEPA.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 31 of its 50 references that are in the map. First posted on arXiv 11 Oct 2024.

Builds on

31 of 50 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  6. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  7. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  8. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  9. 2020Scaling Laws for Neural Language Modelscited by 6,411
  10. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  11. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,853
  12. 2020Exploring Simple Siamese Representation Learningcited by 2,853
19 more
  1. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
  2. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  3. 2022VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Trainingcited by 1,305
  4. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
  5. 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 718
  6. 2024Revisiting Feature Prediction for Learning Visual Representations from Videocited by 490
  7. 2020CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patientscited by 170
  8. 2019Automatic diagnosis of the 12-lead ECG using a deep neural networkcited by 164
  9. 2021Self-supervised Learning is More Robust to Dataset Imbalancecited by 122
  10. 2022Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Languagecited by 108
  11. 2024ECG-FM: An Open Electrocardiogram Foundation Modelcited by 89
  12. 2024Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogramcited by 84
  13. 2021Self-supervised representation learning from 12-lead ECG datacited by 80
  14. 2024Learning by Reconstruction Produces Uninformative Features For Perceptioncited by 54
  15. 2022Lead-agnostic Self-supervised Learning for Local and Global Representations of Electrocardiogramcited by 38
  16. 2020Deep Learning for ECG Segmentationcited by 13
  17. 2023Deep learning based ECG segmentation for delineation of diverse arrhythmiascited by 7
  18. 2023Unsupervised Pre-Training Using Masked Autoencoders for ECG Analysiscited by 7
  19. 2018ECG Segmentation by Neural Networks: Errors and Correctioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 11.

17 papers.

Sort:

By area:

  1. 2025QoQ-Med: Building Multimodal Clinical Foundation Models with Domain-Aware GRPO Trainingcited by 38
  2. 2025CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Modelscited by 13
  3. 2025BenchECG and xECG: a benchmark and baseline for ECG foundation modelscited by 8
  4. 2025Benchmarking ECG FMs: A Reality Check Across Clinical Taskscited by 6
  5. 2025Learning ECG Representations via Poly-Window Contrastive Learningcited by 1
  6. 2025ECG-Soup: Harnessing Multi-Layer Synergy for ECG Foundation Modelscited by 1
  7. 2026Hierarchical Self-Supervised Representation Learning Framework for Multivariate Time Series Grounded in ECG Analysiscited by 1
  8. 2026Extending Pretrained 10-Second ECG Foundation Models to Longer Horizonscited by 0
  9. 2026Information-theoretic Multimodal Representation Learning for Electrocardiogram Signalscited by 0
  10. 2026Do ECG Foundation Models Transfer to Rare Cardiac Diseases? Evidence from Brugada Syndrome Detectioncited by 0
  11. 2026Joint-Embedding Predictive Architecture for Sensor-based Activity Recognitioncited by 0
  12. 2026ECG-InterpBench: Benchmarking the Interpretability of ECG Foundation Models with Matched-Scale Sparse Autoencoderscited by 0
5 more
  1. 2026FOUND-AF: Benchmarking ECG Foundation Models for Atrial Fibrillation Detectioncited by 0
  2. 2026LAEF: A Lead-Agnostic ECG Foundation Model Towards Point-of-Care Diagnosticscited by 0
  3. 2026Capturing Cardiac Cyclicity through Phase-Equivariant Self-Supervised Learningcited by 0
  4. 2026NeuroECG: ECGFounder-Based Deep ECG Representation for EEG-Free Neurological Prognostication After Cardiac Arrestcited by 0
  5. 2026A Hybrid CNN--State-Space--Attention Backbone with Joint-Embedding Predictive Pretraining for 12-Lead ECG Classificationcited by 0