Hybrid CNN–GNN Architecture for Nano-Structured Biomarker Discovery and Risk Scoring in Virtual Preventive Cardiology Systems
Journal of Nano Molecular Intelligence and Virtual Health Systems · 30.06.2025 · 1 (2) · 34
DOI: 10.70023/nmivhs.123 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Nano Molecular Intelligence and Virtual Health Systems, online erschienen am 30.06.2025. Verfasst von Reza Karimi und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Current models for cardiovascular risk prediction are limited in their ability to incorporate imaging signals and the relationships between these imaging signals and biomarker data. Traditional methods do not adequately capture the interdependencies or dynamics of the system. This study introduces a CNN-GNN framework called Nano-GraphFusion for virtual preventive cardiology systems that discovers biomarkers and scores the risk in a nano-structured way. The model combines CNN feature extraction from ECG signals and GNN relation fusion of the biomarkers. The next is the need to have multi-modal learning in clinical datasets, which is not unified. The proposed system enhances the classification accuracy and accuracy in risk stratification. Experimental results demonstrate better predictive power with AUC 0.94 and Macro F1-score 0.91.
Zitieren
APA
Karimi, R., & Hosseini, S. (2025). Hybrid CNN–GNN Architecture for Nano-Structured Biomarker Discovery and Risk Scoring in Virtual Preventive Cardiology Systems. Journal of Nano Molecular Intelligence and Virtual Health Systems, 1(2), 34. https://doi.org/10.70023/nmivhs.123
Vancouver
Karimi R, Hosseini S. Hybrid CNN–GNN Architecture for Nano-Structured Biomarker Discovery and Risk Scoring in Virtual Preventive Cardiology Systems. Journal of Nano Molecular Intelligence and Virtual Health Systems. 2025;1(2):34. doi:10.70023/nmivhs.123
BibTeX
@article{Karimi2025,
author = {Karimi, Reza and Hosseini, Sara},
title = {Hybrid CNN–GNN Architecture for Nano-Structured Biomarker Discovery and Risk Scoring in Virtual Preventive Cardiology Systems},
journal = {Journal of Nano Molecular Intelligence and Virtual Health Systems},
year = {2025},
volume = {1},
number = {2},
pages = {34},
doi = {10.70023/nmivhs.123}
}
EUVIMED