mlsanityData to 7 Oct 2026

An Electrocardiogram Foundation Model Built on over 10 Million Recordings with External Evaluation across Multiple Domains

arXiv 2410.04133 ↗First posted 5 Oct 2024cs.LG cs.AI eess.SPcited by 27

Not on the map. Why?

Abstract (arXiv)

Artificial intelligence (AI) has demonstrated significant potential in ECG analysis and cardiovascular disease assessment. Recently, foundation models have played a remarkable role in advancing medical AI. The development of an ECG foundation model holds the promise of elevating AI-ECG research to new heights. However, building such a model faces several challenges, including insufficient database sample sizes and inadequate generalization across multiple domains. Additionally, there is a notable performance gap between single-lead and multi-lead ECG analyses. We introduced an ECG Foundation Model (ECGFounder), a general-purpose model that leverages real-world ECG annotations from cardiology experts to broaden the diagnostic capabilities of ECG analysis. ECGFounder was trained on over 10 million ECGs with 150 label categories from the Harvard-Emory ECG Database, enabling comprehensive cardiovascular disease diagnosis through ECG analysis. The model is designed to be both an effective out-of-the-box solution, and a to be fine-tunable for downstream tasks, maximizing usability. Importantly, we extended its application to lower rank ECGs, and arbitrary single-lead ECGs in particular. ECGFounder is applicable to supporting various downstream tasks in mobile monitoring scenarios. Experimental results demonstrate that ECGFounder achieves expert-level performance on internal validation sets, with AUROC exceeding 0.95 for eighty diagnoses. It also shows strong classification performance and generalization across various diagnoses on external validation sets. When fine-tuned, ECGFounder outperforms baseline models in demographic analysis, clinical event detection, and cross-modality cardiac rhythm diagnosis. The trained model and data will be publicly released upon publication through the bdsp.io. Our code is available at https://github.com/PKUDigitalHealth/ECGFounder

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). Builds on 13 of its 47 references that are in the map. First posted on arXiv 5 Oct 2024.

Builds on

13 of 47 references.

References · 13

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  3. 2020Designing Network Design Spacescited by 954
  4. 2021Time-Series Representation Learning via Temporal and Contextual Contrastingcited by 342
  5. 2020CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patientscited by 170
  6. 2019Automatic diagnosis of the 12-lead ECG using a deep neural networkcited by 164
  7. 2020Deep Learning for ECG Analysis: Benchmarks and Insights from PTB-XLcited by 158
  8. 2024ECG-FM: An Open Electrocardiogram Foundation Modelcited by 89
  9. 2024Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogramcited by 84
  10. 2023Frozen Language Model Helps ECG Zero-Shot Learningcited by 62
  11. 2020HOLMES: Health OnLine Model Ensemble Serving for Deep Learning Models in Intensive Care Unitscited by 45
  12. 2022Dist-PU: Positive-Unlabeled Learning from a Label Distribution Perspectivecited by 37
1 more
  1. 2024ECG Semantic Integrator (ESI): A Foundation ECG Model Pretrained with LLM-Enhanced Cardiological Textcited by 25

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 10; 2026: 17.

27 papers.

Sort:

  1. 2025OpenECG: Benchmarking ECG Foundation Models with Public 1.2 Million Recordscited by 10
  2. 2025PPGFlowECG: Latent Rectified Flow with Cross-Modal Encoding for PPG-Guided ECG Generation and Cardiovascular Disease Detectioncited by 7
  3. 2025CLEF: Clinically-Guided Contrastive Learning for Electrocardiogram Foundation Modelscited by 7
  4. 2025Self-DANA: A Resource-Efficient Channel-Adaptive Self-Supervised Approach for ECG Foundation Modelscited by 3
  5. 2026Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learningcited by 3
  6. 2025ExChanGeAI: An End-to-End Platform and Efficient Foundation Model for Electrocardiogram Analysis and Fine-tuningcited by 1
  7. 2025BioX-Bridge: Model Bridging for Unsupervised Cross-Modal Knowledge Transfer across Biosignalscited by 1
  8. 2026Cross-Modal Contrastive Learning of ECG and Angiography Representations for Severe Stenosis Classificationcited by 1
  9. 2026P2E-VQ: ECG-linked representation augmentation for PPG via discrete patch retrievalcited by 1
  10. 2025uPVC-Net: A Universal Premature Ventricular Contraction Detection Deep Learning Algorithmcited by 0
  11. 2025Reconstructing 12-Lead ECG from 3-Lead ECG using Variational Autoencoder to Improve Cardiac Disease Detection of Wearable ECG Devicescited by 0
  12. 2025Generalist vs Specialist Time Series Foundation Models: Investigating Potential Emergent Behaviors in Assessing Human Health Using PPG Signalscited by 0
15 more
  1. 2025Modeling Day-Long ECG Signals to Predict Heart Failure Risk with Explainable AIcited by 0
  2. 2026Looking Beyond Accuracy: A Holistic Benchmark of ECG Foundation Modelscited by 0
  3. 2026Task- and Metric-Specific Signal Quality Indices for Medical Time Seriescited by 0
  4. 2026From Foundation ECG Models to NISQ Learners: Distilling ECGFounder into a VQC Studentcited by 0
  5. 2026Bit-Identical Medical Deep Learning via Structured Orthogonal Initializationcited by 0
  6. 2026Domain-Adapted Fine-Tuning of ECG Foundation Models for Multi-Label Structural Heart Disease Screeningcited by 0
  7. 2026Extending Pretrained 10-Second ECG Foundation Models to Longer Horizonscited by 0
  8. 2026How Do Electrocardiogram Models Scale?cited by 0
  9. 2026VitalAgent: A Tool-Augmented Agent for Reactive and Proactive Physiological Monitoring over Wearable Health Datacited by 0
  10. 2026BCG-FM: A Foundation Model for Ambient Cardiac Health Sensingcited by 0
  11. 2026Do ECG Foundation Models Transfer to Rare Cardiac Diseases? Evidence from Brugada Syndrome Detectioncited by 0
  12. 2026LSTrans: Efficient Knowledge Transfer for Lightweight and Automated ECG Classificationcited by 0
  13. 2026The Impact of Temporal Context Length and Encoding Strategies on Self-Supervised ECG Representation Learningcited by 0
  14. 2026Intervention-Aware Clinical World Model for Post-Op Outcome Forecasting in Cardiologycited by 0
  15. 2026Beat-Synchronous Tokenization for ECG Transformerscited by 0