TRACE: Expert-Aligned ECG Representation Learning with Rigorous Benchmarking and Real-World Validation in Acute Cardiac Care

2026-09-28

AI Summary

This study addresses the limitations of existing CLIP-style training under noisy clinical text and the insufficient exploitation of complementary unimodal and cross-modal information by proposing a multimodal ECG representation model. Methodologically, a hybrid architecture is designed to jointly learn unimodal and cross-modal representations. An uncertainty-weighted multi-task learning framework is introduced, leveraging large language models to extract high-fidelity report features for enhanced expert alignment. Experimental results demonstrate that the proposed model achieves robust performance on public benchmarks. Notably, in acute coronary occlusion detection, it yields a 19.0% increase in sensitivity or a 62.6% reduction in false positive rate, significantly outperforming clinical baselines.

Abstract

TRACE (Text-Reinforced Analysis of Cardio ECGs) is a multimodal electrocardiogram (ECG) representation model that learns clinically grounded signal embeddings for downstream cardiac classification. It is designed to address the limitations of existing CLIP-style training, which often struggles with noisy clinical text and fails to leverage the complementary strengths of unimodal (from ECG) and cross-modal (between ECG and matched cardiologist reports) learning. To bridge this gap, we propose a hybrid architecture that jointly learns unimodal and cross-modal representations via uncertainty-weighted multi-task learning while utilizing an LLM-based pipeline to extract high-fidelity findings from cardiologist reports. We evaluate TRACE across a spectrum of clinical urgency, establishing robust performance on public benchmarks for arrhythmia classification and structural abnormalities relative to existing unimodal and multimodal ECG models. To demonstrate real-world utility, we further validate the model on acute coronary occlusion (ACO), where the prevailing ST-elevation criteria miss 25-34% of true occlusions. Utilizing a large private ACO dataset with expert-annotated ground truth, TRACE significantly outperforms real-world clinical practice, yielding a 19.0% increase in sensitivity or a 62.6% reduction in false positive rates at the clinical baseline. This extensive evaluation confirms that TRACE delivers both strong performance on benchmark tasks and tangible clinical impact in the most acute, high-risk cardiac scenarios.

Problem

• ECG representation learning
• multimodal learning
• noisy clinical text
• acute coronary occlusion
• clinical benchmarking

Innovation

• Multimodal ECG Representation Learning
• Uncertainty-Weighted Multi-Task Learning
• LLM-based Clinical Text Extraction
• Acute Coronary Occlusion Detection
• Cross-Modal Alignment