Demographic-Aware Self-Supervised Anomaly Detection Pretraining for Equitable Rare Cardiac Diagnosis

2026-03-20

AI Summary

This work addresses the diagnostic challenges posed by rare electrocardiogram (ECG) abnormalities, which are often underrepresented and unevenly distributed across demographic groups, leading to detection difficulties and inequitable diagnosis. The authors propose a two-stage AI framework for ECG analysis: first, a self-supervised pretraining phase that integrates masked signal reconstruction, trend modeling, and demographic attribute prediction to learn robust and equitable ECG representations; followed by multi-label fine-tuning with an asymmetric loss and the generation of abnormality heatmaps for precise localization. Notably, this is the first approach to incorporate demographic awareness into self-supervised ECG pretraining. Evaluated on over one million ECGs, the method achieves a 94.7% AUROC, reduces the performance gap between rare and common abnormalities by 73%, and maintains consistent accuracy across age and sex subgroups, substantially enhancing both diagnostic fairness and clinical utility.

Technology Category

• Machine Learning: Ethics, Bias, and Fairness
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Computer Vision: Bias, Fairness & Privacy

Application Category

• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Rare cardiac anomalies are difficult to detect from electrocardiograms (ECGs) due to their long-tailed distribution with extremely limited case counts and demographic disparities in diagnostic performance. These limitations contribute to delayed recognition and uneven quality of care, creating an urgent need for a generalizable framework that enhances sensitivity while ensuring equity across diverse populations. In this study, we developed an AI-assisted two-stage ECG framework integrating self-supervised anomaly detection with demographic-aware representation learning. The first stage performs self-supervised anomaly detection pretraining by reconstructing masked global and local ECG signals, modeling signal trends, and predicting patient attributes to learn robust ECG representations without diagnostic labels. The pretrained model is then fine-tuned for multi-label ECG classification using asymmetric loss to better handle long-tail cardiac abnormalities, and additionally produces anomaly score maps for localization, with CPU-based optimization enabling practical deployment. Evaluated on a longitudinal cohort of over one million clinical ECGs, our method achieves an AUROC of 94.7% for rare anomalies and reduces the common-rare performance gap by 73%, while maintaining consistent diagnostic accuracy across age and sex groups. In conclusion, the proposed equity-aware AI framework demonstrates strong clinical utility, interpretable anomaly localization, and scalable performance across multiple cohorts, highlighting its potential to mitigate diagnostic disparities and advance equitable anomaly detection in biomedical signals and digital health. Source code is available at https://github.com/MediaBrain-SJTU/Rare-ECG.

Problem

• rare cardiac anomalies
• long-tailed distribution
• demographic disparities
• equitable diagnosis
• ECG anomaly detection

Innovation

• self-supervised learning
• demographic-aware representation
• anomaly detection
• long-tailed classification
• equitable AI