EchoGraph-FusionNet: Adaptive Disease-Graph and Mixture-of-Experts Learning for Predicting Echocardiographic Abnormalities from ECGs
Life · 25.10.2026 · 16 (11) · 1795
DOI: 10.3390/life16111795 ↗Kurz gefasst
Fachartikel in der Zeitschrift Life, online erschienen am 25.10.2026. Verfasst von Çetin Mirzaoğlu und 4 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Firat University Scientific Research Project Management Unit.
Abstract (Originalsprache)
Background/Objectives: Cardiovascular diseases remain a major global health burden, while echocardiography, despite its diagnostic value, is not always readily accessible. This study proposes EchoGraph-FusionNet, a multimodal deep learning framework for predicting echocardiographic abnormalities and structural heart disease (SHD) from 12-lead electrocardiograms (ECGs) and clinical variables.
Methods
The framework integrates ECG and clinical representations through cross-modal fusion, followed by a disease-aware mixture-of-experts module and a two-layer adaptive graph attention network that jointly models 11 echocardiographic abnormalities. Five continuous echocardiographic measurements were incorporated as auxiliary regression tasks. Prevalence-aware logit adjustment and asymmetric loss were employed to address class imbalance. Performance was evaluated on the predefined EchoNext training, validation, and held-out test partitions across three independent random seeds.
Results
On the held-out test set, EchoGraph-FusionNet achieved a macro AUROC of 0.8066 ± 0.0025 and macro AUPRC of 0.2838 ± 0.0056 across 11 abnormalities. For SHD detection, AUROC and AUPRC reached 0.8345 ± 0.0001 and 0.8039 ± 0.0031, respectively.
Conclusions
EchoGraph-FusionNet effectively combines multimodal representations, disease relationships, and multitask learning for ECG-based prediction of echocardiographic abnormalities, supporting its potential as a screening and prioritization approach for structural cardiac assessment.
Förderung
- Firat University Scientific Research Project Management Unit
Zitieren
APA
Mirzaoğlu, Ç., Çakmak, T., Demir, M., Demir, F., & Alçin, Ö. F. (2026). EchoGraph-FusionNet: Adaptive Disease-Graph and Mixture-of-Experts Learning for Predicting Echocardiographic Abnormalities from ECGs. Life, 16(11), 1795. https://doi.org/10.3390/life16111795
Vancouver
Mirzaoğlu Ç, Çakmak T, Demir M, Demir F, Alçin ÖF. EchoGraph-FusionNet: Adaptive Disease-Graph and Mixture-of-Experts Learning for Predicting Echocardiographic Abnormalities from ECGs. Life. 2026;16(11):1795. doi:10.3390/life16111795
BibTeX
@article{Mirzaolu2026,
author = {Mirzaoğlu, Çetin and Çakmak, Tolga and Demir, Melike and Demir, Fatih and Alçin, Ömer Faruk},
title = {EchoGraph-FusionNet: Adaptive Disease-Graph and Mixture-of-Experts Learning for Predicting Echocardiographic Abnormalities from ECGs},
journal = {Life},
year = {2026},
volume = {16},
number = {11},
pages = {1795},
doi = {10.3390/life16111795}
}
EUVIMED