Domain Knowledge is Power: Leveraging Physiological Priors for Self Supervised Representation Learning in Electrocardiography

2025-09-09

AI Summary

To address the limited generalizability and clinical relevance of ECG arrhythmia classification models caused by scarce labeled data, this paper proposes PhysioCLR—a physiology-informed self-supervised contrastive learning framework. Methodologically, it (1) constructs physiology-aware positive/negative sample pairs incorporating ECG waveform similarity constraints; (2) designs class-preserving data augmentation to retain diagnostically critical features; and (3) employs a hybrid loss function jointly optimizing representation discriminability and clinical interpretability. Evaluated on the Chapman, Georgia, and a private ICU dataset, PhysioCLR achieves an average AUROC improvement of 12% over the strongest baseline, demonstrating significantly enhanced cross-center generalization. This work establishes a novel paradigm for trustworthy ECG analysis in low-resource clinical settings.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Objective: Electrocardiograms (ECGs) play a crucial role in diagnosing heart conditions; however, the effectiveness of artificial intelligence (AI)-based ECG analysis is often hindered by the limited availability of labeled data. Self-supervised learning (SSL) can address this by leveraging large-scale unlabeled data. We introduce PhysioCLR (Physiology-aware Contrastive Learning Representation for ECG), a physiology-aware contrastive learning framework that incorporates domain-specific priors to enhance the generalizability and clinical relevance of ECG-based arrhythmia classification. Methods: During pretraining, PhysioCLR learns to bring together embeddings of samples that share similar clinically relevant features while pushing apart those that are dissimilar. Unlike existing methods, our method integrates ECG physiological similarity cues into contrastive learning, promoting the learning of clinically meaningful representations. Additionally, we introduce ECG- specific augmentations that preserve the ECG category post augmentation and propose a hybrid loss function to further refine the quality of learned representations. Results: We evaluate PhysioCLR on two public ECG datasets, Chapman and Georgia, for multilabel ECG diagnoses, as well as a private ICU dataset labeled for binary classification. Across the Chapman, Georgia, and private cohorts, PhysioCLR boosts the mean AUROC by 12% relative to the strongest baseline, underscoring its robust cross-dataset generalization. Conclusion: By embedding physiological knowledge into contrastive learning, PhysioCLR enables the model to learn clinically meaningful and transferable ECG eatures. Significance: PhysioCLR demonstrates the potential of physiology-informed SSL to offer a promising path toward more effective and label-efficient ECG diagnostics.

Problem

• Addressing limited labeled ECG data for AI-based heart condition diagnosis
• Enhancing ECG representation learning with physiological domain knowledge
• Improving cross-dataset generalization in arrhythmia classification tasks

Innovation

• Integrates physiological priors into contrastive learning
• Uses ECG-specific augmentations preserving clinical categories
• Proposes hybrid loss function refining representation quality