EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction

Huiya Zhao iD, Dehao Sui, Yasha Wang, Liantao Ma, Ling Wang

Sensors · 09.04.2025 · 25 (8) · 2374

DOI: 10.3390/s25082374 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 09.04.2025. Verfasst von Huiya Zhao und 4 weiteren Autor:innen. Beteiligte Einrichtung: National Engineering Research Center for Software Engineering, Peking University, Beijing 100871, China. Hauptthemen laut MEDLINE: Electronic Health Records, Privacy. Die Arbeit wurde bisher 42-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Xuzhou Scientific Technological Projects, the China Postdoctoral Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Secure and privacy-preserving health status representation learning has become a critical challenge in clinical prediction systems. While deep learning models require substantial high-quality data for training, electronic health records are often restricted by strict privacy regulations and institutional policies, particularly during emerging health crises. Traditional approaches to data integration across medical institutions face significant privacy and security challenges, as healthcare providers cannot directly share patient data. This work presents MultiProg, a secure federated learning framework for clinical representation learning. Our approach enables multiple medical institutions to collaborate without exchanging raw patient data, maintaining data locality while improving model performance. The framework employs a multi-channel architecture where institutions share only the low-level feature extraction layers, protecting sensitive patient information. We introduce a feature calibration mechanism that ensures robust performance even with heterogeneous feature sets across different institutions. Through extensive experiments, we demonstrate that the framework successfully enables secure knowledge sharing across institutions without compromising sensitive patient data, achieving enhanced predictive capabilities compared to isolated institutional models. Compared to state-of-the-art methods, our approach achieves the best performance across multiple datasets with statistically significant improvements.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Electronic Health RecordsPrivacyHumansComputer SecurityPrognosisDeep LearningMachine LearningAlgorithmsConfidentialityFederated Learning

Stichwörter der Autor:innen

federated learninghealthcare privacymulti-institutional collaboration

Autor:innen und Institutionen

  • National Engineering Research Center for Software Engineering, Peking University, Beijing 100871, China.
  • Key Laboratory of High Confidence Software Technologies, Ministry of Education, Beijing 100871, China.
  • Affiliated Xuzhou Municipal Hospital of Xuzhou Medical University, Xuzhou 221002, China.

ORCID-Profile: Huiya Zhao

Förderung

  • National Natural Science Foundation of China
  • Xuzhou Scientific Technological Projects
  • the China Postdoctoral Science Foundation

Zitieren

APA

Zhao, H., Sui, D., Wang, Y., Ma, L., & Wang, L. (2025). Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction. Sensors, 25(8), 2374. https://doi.org/10.3390/s25082374

Vancouver

Zhao H, Sui D, Wang Y, Ma L, Wang L. Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction. Sensors. 2025;25(8):2374. doi:10.3390/s25082374

BibTeX
@article{Zhao2025,
  author = {Zhao, Huiya and Sui, Dehao and Wang, Yasha and Ma, Liantao and Wang, Ling},
  title = {Privacy-Preserving Federated Learning Framework for Multi-Source Electronic Health Records Prognosis Prediction},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {8},
  pages = {2374},
  doi = {10.3390/s25082374}
}
Kostenlos starten →