Foundation Model for Cardiac Time Series via Masked Latent Attention

TL;DR

This paper introduces LAMAE, a foundation model for ECG analysis that leverages cross-lead structural redundancy through latent attention, enhancing representation quality and transferability.

Contribution

The novel LAMAE model explicitly models cross-lead interactions in ECGs using latent attention, improving self-supervised pretraining for cardiovascular diagnosis.

Findings

LAMAE outperforms baseline models in ICD-10 code prediction.

Leveraging cross-lead structure improves ECG representation quality.

Empirical results on Mimic-IV-ECG demonstrate the effectiveness of the approach.

Abstract

Electrocardiograms (ECGs) are among the most widely available clinical signals and play a central role in cardiovascular diagnosis. While recent foundation models (FMs) have shown promise for learning transferable ECG representations, most existing pretraining approaches treat leads as independent channels and fail to explicitly leverage their strong structural redundancy. We introduce the latent attention masked autoencoder (LAMAE) FM that directly exploits this structure by learning cross-lead connection mechanisms during self-supervised pretraining. Our approach models higher-order interactions across leads through latent attention, enabling permutation-invariant aggregation and adaptive weighting of lead-specific representations. We provide empirical evidence on the Mimic-IV-ECG database that leveraging the cross-lead connection constitutes an effective form of structural supervision, improving representation quality and transferability. Our method shows strong performance in predicting ICD-10 codes, outperforming independent-lead masked modeling and alignment-based baselines.