Federated Graph Learning Enables Decentralized Disease Surveillance Across Health Systems
Academic Journal of Applied Sciences · 02.04.2026 · 1 (1) · 95-103
DOI: 10.54097/r1bpev97 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Journal of Applied Sciences, online erschienen am 02.04.2026. Verfasst von Luca Moretti und 1 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The emergence of federated graph learning (FGL) offers a transformative paradigm for disease surveillance across distributed health systems, enabling collaborative model training without centralizing sensitive patient records. Conventional disease monitoring pipelines require aggregating clinical data into unified repositories, which raises significant concerns related to patient privacy, regulatory compliance, and institutional data governance. This paper presents a decentralized surveillance framework that integrates graph neural network (GNN) architectures with federated optimization protocols, allowing geographically dispersed hospitals, public health agencies, and community clinics to jointly learn disease propagation dynamics from locally retained data. The proposed system constructs patient-population interaction graphs at each participating node and trains a spatiotemporal GNN under a privacy-preserving federated aggregation scheme. Empirical evaluations using simulated multi-site infectious disease datasets demonstrate that the federated model achieves detection accuracy within 3.2% of a centralized training baseline while substantially reducing inter-institutional data exposure. This work advances the intersection of distributed machine learning and public health informatics, providing a principled and scalable foundation for real-time epidemic intelligence systems that respect institutional data sovereignty.
Zitieren
APA
Moretti, L., & Kapoor, A. (2026). Federated Graph Learning Enables Decentralized Disease Surveillance Across Health Systems. Academic Journal of Applied Sciences, 1(1), 95-103. https://doi.org/10.54097/r1bpev97
Vancouver
Moretti L, Kapoor A. Federated Graph Learning Enables Decentralized Disease Surveillance Across Health Systems. Academic Journal of Applied Sciences. 2026;1(1):95-103. doi:10.54097/r1bpev97
BibTeX
@article{Moretti2026,
author = {Moretti, Luca and Kapoor, Ananya},
title = {Federated Graph Learning Enables Decentralized Disease Surveillance Across Health Systems},
journal = {Academic Journal of Applied Sciences},
year = {2026},
volume = {1},
number = {1},
pages = {95--103},
doi = {10.54097/r1bpev97}
}
EUVIMED