Skip to content
Preprint

Graph-Based Pseudo-multimodal Contrastive Learning for 12-Lead ECG Representations

Aug 2026 · 0 citations · 11 references
Computer Science

TL;DR

Results suggest the effectiveness of using GADF as a complementary representation and incorporating explicit graph-based modeling of inter-lead dependency for learning robust 12-lead ECG representations, and suggest that the proposed framework achieves competitive performance compared to supervised learning methods.

Abstract

12-lead electrocardiogram (ECG) is a standard, non-invasive examination widely used for diagnosing coronary artery disease, where clinical interpretation relies on comparing waveform patterns across multiple leads. However, most existing ECG analysis methods focus on single-lead signals or treat each lead independently, and typically process ECG signals as one-dimensional time-series data using CNNs or RNNs. While effective in modeling local waveform changes, such approaches have difficulty capturing inter-lead dependency and global waveform patterns essential for clinical diagnosis. To address this limitation, we propose a graph-based pseudo-multimodal contrastive learning framework called Graph-CMMC. ECG waveforms are transformed into Gramian Angular Difference Field (GADF) images to construct complementary representations of the same cardiac activity, enabling a pseudo-multimodal learning setting. Using all 12 leads, Graph-CMMC aligns waveform and GADF representations in a self-supervised manner, while a graph-based relational module is employed to model inter-lead dependency and enforce structural consistency across leads during contrastive learning. Experimental results on a multi-label coronary artery occlusion classification task demonstrate that the proposed framework achieves competitive performance compared to supervised learning methods. These results further suggest the effectiveness of using GADF as a complementary representation and incorporating explicit graph-based modeling of inter-lead dependency for learning robust 12-lead ECG representations.

View source

Similar papers

Open access Aug 2026

Inter-Patient Multi-Label Localization of Myocardial Infarction Territories Using 12-Lead ECG Signals

The localization of Myocardial Infarction (MI) from Electrocardiogram (ECG) signals remains challenging due to inter-patient variability, class imbalance, and limited interpretability of current methods. This study presents a one-dimensional convolutional network, MobileECGNet, for multi-label, territory-wise MI locali...

Ankita Shukla, Muhammed Izharuddin · 0 citations
Open access Aug 2026

A deformable lead-attention fusion network for multi-label ECG classification integrating clinical metadata

Cardiovascular diseases remain a major global health burden, making accurate electrocardiogram (ECG) analysis essential for timely diagnosis. While deep learning-based methods have made significant progress in automating ECG classification, they are often limited by the variability of ECG signals, the frequent presence...

En-Jun Zhou, Xiang-Yong Kong, Hao Wang et al. · 0 citations
Open access Oct 2026

12-lead ECG synthesis from reduced lead set using multi-lead convolutional autoencoder

Abstract Accurate reconstruction of the standard 12-lead electrocardiogram (ECG) from a reduced number of leads has significant potential for simplifying cardiac monitoring systems and enabling wearable healthcare solutions. This study proposes a novel multi-lead convolutional autoencoder (MLCAE) for reconstructing sta...

F. Samann, T. Schanze · 0 citations
Sep 2026

X-Beat: An Explainable Framework for ECG Image Classification

X- Beat is presented, an explainable and reliability-aware benchmark framework for ECG image classification designed to support trustworthy AI systems in healthcare and provides a structured and reproducible bench- mark for evaluating both predictive performance and explanation reliability in ECG image classification.

Mohammad Sadman Tahsin, Haitham Y. Adarbah, A. Noore · 0 citations

Signals Are All You Need: A Deep Learning Architecture for ECG Lead Classification

A permutation-equivariant deep learning architecture that infers lead labels directly from waveforms, using convolutional, recurrent, and attention components to model temporal and inter-lead structure is proposed.

Nathan Macdonald, R. Peeters, P. Bonizzi et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.