Abstract: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 <a class="link-external link-http" href="http://bdsp.io" rel="external noopener nofollow">this http URL</a>. Our code is available at <a class="link-external link-https" href="https://github.com/bdsp-core/ECGFounder" rel="external noopener nofollow">this https URL</a>

A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients

An Open-Access Arrhythmia Database of Wearable Electrocardiogram

[A Study on Clustering Analysis of Arrhythmias].

Ecglens: Interactive Visual Exploration Of Large Scale Ecg Data For Arrhythmia Detection

Chinese Cardiovascular Disease Database (CCDD) and Its Management Tool

The Harvard-Emory ECG Database

Ccdd: An Enhanced Standard Ecg Database With Its Management And Annotation Tools

A Novel Lightweight Computerized ECG Interpretation Approach Based on Clinical 12-Lead Data

An Open Access Database for Evaluating the Algorithms of Electrocardiogram Rhythm and Morphology Abnormality Detection

An exercise ECG database with synchronized exercise information

Inter-patient Heartbeat Classification Based on Region Feature Extraction and Ensemble Classifier

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

A community effort to assess and improve computerized interpretation of 12-lead resting electrocardiogram

Arrhythmia classification detection based on multiple electrocardiograms databases

Survey on atrial fibrillation detection from a single-lead ECG wave for Internet of Medical Things

rECGnition_v1.0: Arrhythmia detection using cardiologist-inspired multi-modal architecture incorporating demographic attributes in ECG

A cascaded classifier for multi-lead ECG based on feature fusion

Cardiologist-level interpretable knowledge-fused deep neural network for automatic arrhythmia diagnosis

Feature Engineering and Computational Intelligence in ECG Monitoring

Intelligent Electrocardiogram Analysis in Medicine: Data, Methods, and Applications.

ECG-Image-Database: A Dataset of ECG Images with Real-World Imaging and Scanning Artifacts; A Foundation for Computerized ECG Image Digitization and Analysis