Causal Physiological Representation Learning for Robust ECG Analysis
Research Paper#Medical AI, ECG Analysis, Adversarial Robustness, Causal Inference🔬 Research|Analyzed: Jan 3, 2026 09:18•
Published: Dec 31, 2025 02:08
•1 min read
•ArXivAnalysis
This paper addresses the vulnerability of deep learning models for ECG diagnosis to adversarial attacks, particularly those mimicking biological morphology. It proposes a novel approach, Causal Physiological Representation Learning (CPR), to improve robustness without sacrificing efficiency. The core idea is to leverage a Structural Causal Model (SCM) to disentangle invariant pathological features from non-causal artifacts, leading to more robust and interpretable ECG analysis.
Key Takeaways
- •Proposes CPR, a novel method for robust ECG analysis.
- •CPR uses a Structural Causal Model (SCM) to disentangle causal and non-causal features.
- •CPR outperforms existing methods in robustness against adversarial attacks while maintaining efficiency.
- •CPR offers a superior trade-off between robustness, efficiency, and clinical interpretability.
Reference / Citation
View Original"CPR achieves an F1 score of 0.632 under SAP attacks, surpassing Median Smoothing (0.541 F1) by 9.1%."