mlsanityData to 14 Oct 2026

Reading Your Heart: Learning ECG Words and Sentences via Pre-training ECG Language Model

arXiv 2502.10707 ↗First posted 15 Feb 2025cs.LG cs.AIcited by 36

Not on the map. Why?

Abstract (arXiv)

Electrocardiogram (ECG) is essential for the clinical diagnosis of arrhythmias and other heart diseases, but deep learning methods based on ECG often face limitations due to the need for high-quality annotations. Although previous ECG self-supervised learning (eSSL) methods have made significant progress in representation learning from unannotated ECG data, they typically treat ECG signals as ordinary time-series data, segmenting the signals using fixed-size and fixed-step time windows, which often ignore the form and rhythm characteristics and latent semantic relationships in ECG signals. In this work, we introduce a novel perspective on ECG signals, treating heartbeats as words and rhythms as sentences. Based on this perspective, we first designed the QRS-Tokenizer, which generates semantically meaningful ECG sentences from the raw ECG signals. Building on these, we then propose HeartLang, a novel self-supervised learning framework for ECG language processing, learning general representations at form and rhythm levels. Additionally, we construct the largest heartbeat-based ECG vocabulary to date, which will further advance the development of ECG language processing. We evaluated HeartLang across six public ECG datasets, where it demonstrated robust competitiveness against other eSSL methods. Our data and code are publicly available at https://github.com/PKUDigitalHealth/HeartLang.

Key facts. Cited by 36 papers in the map (data through 14 Oct 2026). Builds on 22 of its 37 references that are in the map. First posted on arXiv 15 Feb 2025.

Builds on

22 of 37 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  4. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  5. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  6. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,616
  7. 2020On Layer Normalization in the Transformer Architecturecited by 922
  8. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 345
  9. 2024Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCIcited by 327
  10. 2022BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizerscited by 303
  11. 2020CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patientscited by 171
  12. 2020Deep Learning for ECG Analysis: Benchmarks and Insights from PTB-XLcited by 158
10 more
  1. 2022Unsupervised Time-Series Representation Learning with Iterative Bilinear Temporal-Spectral Fusioncited by 96
  2. 2024Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogramcited by 84
  3. 2024Zero-Shot ECG Classification with Multimodal Learning and Test-time Clinical Knowledge Enhancementcited by 67
  4. 2022Self-Supervised Time Series Representation Learning via Cross Reconstruction Transformercited by 47
  5. 2022Lead-agnostic Self-supervised Learning for Local and Global Representations of Electrocardiogramcited by 38
  6. 2021Intra-Inter Subject Self-supervised Learning for Multivariate Cardiac Signalscited by 35
  7. 2019MINA: Multilevel Knowledge-Guided Attention for Modeling Electrocardiography Signalscited by 21
  8. 2023ECGBERT: Understanding Hidden Language of ECGs with Self-Supervised Representation Learningcited by 12
  9. 2020ECG Language Processing (ELP): a New Technique to Analyze ECG Signalscited by 7
  10. 2022A Simple Self-Supervised ECG Representation Learning Method via Manipulated Temporal-Spatial Reverse Detectioncited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 12; 2026: 24.

36 papers.

Sort:

By area:

  1. 2025CodeBrain: Bridging Decoupled Tokenizer and Multi-Scale Architecture for EEG Foundation Modelcited by 28
  2. 2025From Token to Rhythm: A Multi-Scale Approach for ECG-Language Pretrainingcited by 16
  3. 2025Towards Robust Multimodal Physiological Foundation Models: Handling Arbitrary Missing Modalitiescited by 11
  4. 2026ECG-R1: Protocol-Guided and Modality-Agnostic MLLM for Reliable ECG Interpretationcited by 9
  5. 2025CLEF: Clinically-Guided Contrastive Learning for Electrocardiogram Foundation Modelscited by 7
  6. 2025SuPreME: A Supervised Pre-training Framework for Multimodal ECG Representation Learningcited by 6
  7. 2025Knowledge-enhanced Multimodal ECG Representation Learning with Arbitrary-Lead Inputscited by 5
  8. 2025UniECG: Understanding and Generating ECG in One Unified Modelcited by 4
  9. 2025Self-Alignment Learning to Improve Myocardial Infarction Detection from Single-Lead ECGcited by 4
  10. 2025Signal, Image, or Symbolic: Exploring the Best Input Representation for Electrocardiogram-Language Models Through a Unified Frameworkcited by 3
  11. 2025HeartcareGPT: A Unified Multimodal ECG Suite for Dual Signal-Image Modeling and Understandingcited by 3
  12. 2026RhythmBERT: A Self-Supervised Language Model Based on Latent Representations of ECG Waveforms for Heart Disease Detectioncited by 3
24 more
  1. 2026CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representationcited by 3
  2. 2025Tracing the Heart's Pathways: ECG Representation Learning from a Cardiac Conduction Perspectivecited by 2
  3. 2026Foundation Model for Cardiac Time Series via Masked Latent Attentioncited by 1
  4. 2026How Do Electrocardiogram Models Scale?cited by 1
  5. 2026LVCG: Learning ECG Representations in the Latent Vectorcardiogram Spacecited by 1
  6. 2025Approaching Low-Cost Cardiac Intelligence with Semi-Supervised Knowledge Distillationcited by 0
  7. 2026BEAT-Net: Injecting Biomimetic Spatio-Temporal Priors for Interpretable ECG Diagnosiscited by 0
  8. 2026ELF: A Family of Encoder-Free ECG-Language Modelscited by 0
  9. 2026Position: Evaluation of ECG Representations Must Be Fixedcited by 0
  10. 2026Learning ECG Image Representations via Dual Physiological-Aware Alignmentscited by 0
  11. 2026PanLUNA: An Efficient and Robust Query-Unified Multimodal Model for Edge Biosignal Intelligencecited by 0
  12. 2026CoRe-ECG: Advancing Self-Supervised Representation Learning for 12-Lead ECG via Contrastive and Reconstructive Synergycited by 0
  13. 2026From Muscle Bursts to Motor Intent: Self-Supervised Token Modeling for Heterogeneous EMGcited by 0
  14. 2026Reasoning Before Diagnosis: Physician-Inspired Structured Thinking for ECG Classificationcited by 0
  15. 2026Information-theoretic Multimodal Representation Learning for Electrocardiogram Signalscited by 0
  16. 2026MyoSem: Aligning Electromyography with Natural-Language Action Semanticscited by 0
  17. 2026SPOTR: Spatio-temporal Pooling One-Token Reconstruction for Universal Physiological Signal Self-supervised Learningcited by 0
  18. 2026EchoBridge: Long-Tail-Aware ECG-Echocardiography Text Alignment for Echocardiography-Derived Cardiac Findingscited by 0
  19. 2026LAEF: A Lead-Agnostic ECG Foundation Model Towards Point-of-Care Diagnosticscited by 0
  20. 2026ECG-LENS: Lead-Aware Clinical Context Enriched ECG Report Generation and Evaluationcited by 0
  21. 2026The Impact of Temporal Context Length and Encoding Strategies on Self-Supervised ECG Representation Learningcited by 0
  22. 2026Beat-Synchronous Tokenization for ECG Transformerscited by 0
  23. 2026Reading the Whole Heart: Latent-Attention Masked Autoencoders for Multimodal Cardiac Representation Learningcited by 0
  24. 2026TRACE: Tractable Routing Autoencoder for Clinical ECGcited by 0