Boosting Masked ECG-Text Auto-Encoders as Discriminative Learners

TL;DR

D-BETA is a novel contrastive masked auto-encoder framework that enhances cross-modal ECG-text representations, significantly improving diagnostic performance with limited data and zero-shot capabilities.

Contribution

The paper introduces D-BETA, a new pre-training method combining generative and discriminative learning for ECG-text data, addressing modality disparities and data scarcity.

Findings

Achieves 15% AUC improvement with 1% training data

Outperforms existing methods in zero-shot performance by 2%

Demonstrates robustness across five public datasets

Abstract

The accurate interpretation of Electrocardiogram (ECG) signals is pivotal for diagnosing cardiovascular diseases. Integrating ECG signals with accompanying textual reports further holds immense potential to enhance clinical diagnostics by combining physiological data and qualitative insights. However, this integration faces significant challenges due to inherent modality disparities and the scarcity of labeled data for robust cross-modal learning. To address these obstacles, we propose D-BETA, a novel framework that pre-trains ECG and text data using a contrastive masked auto-encoder architecture. D-BETA uniquely combines the strengths of generative with boosted discriminative capabilities to achieve robust cross-modal representations. This is accomplished through masked modality modeling, specialized loss functions, and an improved negative sampling strategy tailored for cross-modal alignment. Extensive experiments on five public datasets across diverse downstream tasks demonstrate that D-BETA significantly outperforms existing methods, achieving an average AUC improvement of 15% in linear probing with only one percent of training data and 2% in zero-shot performance without requiring training data over state-of-the-art models. These results highlight the effectiveness of D-BETA, underscoring its potential to advance automated clinical diagnostics through multi-modal representations. Our sample code and checkpoint are made available at https://github.com/manhph2211/D-BETA.