Organizations: Graduate School of Engineering Science Yokohama National University Yokohama, Japan · Division of Cardiology Yokohama City University Medical Center Yokohama, Japan · Technology and Innovation Department Fukuda Denshi Co.,Ltd Tokyo, Japan
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.
Figures & tables
Fig. 1: An overview of Graph-CMMC
Fig. 2: An overview of GADF Transform
Fig. 3: An overview of CMLC
Fig. 4: An overview of CMMC
Fig. 5: An overview of CMMC+CMLC
Method
Macro F1
Subset Accuracy
Overall Accuracy
Jaccard
Supervised baseline
0.822
0.771
0.903
0.786
CMLC
0.563
0.432
0.797
0.515
CMLC+CMMC
0.703
0.593
0.846
0.628
CMMC
0.729
0.627
0.857
0.665
Graph-CMMC (d1)
0.746
0.670
0.855
0.715
Graph-CMMC (d2)
0.776
0.711
0.873
0.752
TABLE I: Experimental results
Model
LAD
LCX
LMT
RCA
Supervised baseline
0.781
0.809
0.936
0.763
CMLC
0.328
0.526
0.769
0.629
CMLC+CMMC
0.680
0.602
0.834
0.696
CMMC
0.726
0.607
0.859
0.723
Graph-CMMC (d1)
0.739
0.697
0.861
0.686
Graph-CMMC (d2)
0.758
0.720
0.880
0.745
TABLE II: Per-class F1-score
Method
Mean ( ↓ )
Std ( ↓ )
Min
Max
CMMC
0.0659
0.0265
0.0188
0.1190
CMLC
0.0648
0.0335
0.0145
0.1735
CMMC+CMLC
0.0639
0.0264
0.0150
0.1381
Graph-CMMC
0.0589
0.0265
0.0119
0.1248
TABLE III: Inter-lead embedding distance statistics for different methods
Fig. 6: Boxplot comparison of inter-lead distance distributions for different methods
Fig. 7: Visualization of the inter-lead dependency learned by Graph-CMMC. (a) Heatmap of the learned adjacency matrix. (b) Louvain clustering results based on the learned adjacency matrix.
Accurate interpretation of electrocardiogram (ECG) signals is crucial for diagnosing cardiovascular diseases. Recent multimodal approaches that integrate ECGs with accompanying clinical reports show strong potential, but they still face two main concerns from a modality perspective: (1) intra-modality: existing models process ECGs in a lead-agnostic manner, overlooking spatial-temporal dependencies across leads, which restricts their effectiveness in modeling fine-grained diagnostic patterns; (2) inter-modality: existing methods directly align ECG signals with clinical reports, introducing modality-specific biases due to the free-text nature of the reports. In light of these two issues, we propose CG-DMER, a contrastive-generative framework for disentangled multimodal ECG representation learning, powered by two key designs: (1) Spatial-temporal masked modeling is designed to better capture fine-grained temporal dynamics and inter-lead spatial dependencies by applying masking across both spatial and temporal dimensions and reconstructing the missing information. (2) A representation disentanglement and alignment strategy is designed to mitigate unnecessary noise and modality-specific biases by introducing modality-specific and modality-shared encoders, ensuring a clearer separation between modality-invariant and modality-specific representations. Experiments on three public datasets demonstrate that CG-DMER achieves state-of-the-art performance across diverse downstream tasks.
Ziwei Niu, Hao Sun, Shujun Bian +4
Department of Biomedical Engineering, National University of Singapore, Singapore, Singapore · College of Computer Science and Technology, Zhejiang University, Hangzhou, China · College of information science and engineering, Ritsumeikan university, ibaraki Osaka, Japan +1
Reconstructing a 12-lead electrocardiogram (ECG) from a reduced lead set is an ill-posed inverse problem due to anatomical variability. Standard deep learning methods often ignore underlying cardiac pathology losing vital morphology in precordial leads. We propose Pathology-Aware Multi-View Contrastive Learning, a framework that regularizes the latent space through a pathological manifold. Our architecture integrates high-fidelity time-domain waveforms with pathology-aware embeddings learned via supervised contrastive alignment. By maximizing mutual information between latent representations and clinical labels, the framework learns to filter anatomical "nuisance" variables. On the PTB-XL dataset, our method achieves approx. 76% reduction in RMSE compared to state-of-the-art model in patient-independent setting. Cross-dataset evaluation on the PTB Diagnostic Database confirms superior generalization, bridging the gap between hardware portability and diagnostic-grade reconstruction.
Youssef Youssef, Jitin Singla
Department of Computer Science and Engineering, Indian Institute of Technology Roorkee, Roorkee 247667, India · Department of Biosciences and Bioengineering, Indian Institute of Technology Roorkee, Roorkee 247667, India
Electrocardiograms (ECGs) are among the most widely used diagnostic tools for cardiovascular diseases, and a large amount of ECG data worldwide appears only in image form. However, most existing automated ECG analysis methods rely on access to raw signal recordings, limiting their applicability in real-world and resource-constrained settings. In this paper, we present ECG-Scan, a self-supervised framework for learning clinically generalized representations from ECG images through dual physiological-aware alignments: 1) Our approach optimizes image representation learning using multimodal contrastive alignment between image and gold-standard signal-text modalities. 2) We further integrate domain knowledge via soft-lead constraints, regularizing the reconstruction process and improving signal lead inter-consistency. Extensive benchmarking across multiple datasets and downstream tasks demonstrates that our image-based model achieves superior performance compared to existing image baselines and notably narrows the gap between ECG image and signal analysis. These results highlight the potential of self-supervised image modeling to unlock large-scale legacy ECG data and broaden access to automated cardiovascular diagnostics.
Hung Manh Pham, Jialu Tang, Aaqib Saeed +3
University of Cambridge · Singapore Management University · Eindhoven University of Technology, Netherlands