Leveraging Federated Learning Models to Enhance Privacy-Preserving Predictive Analytics in Distributed Healthcare Information Systems

Abstract

The proliferation of interconnected healthcare systems has enabled the collection and processing of vast amounts of patient data for predictive analytics. However, data privacy remains a paramount concern, especially in distributed environments. This paper explores how federated learning (FL) offers a decentralized solution by enabling collaborative model training without the need to share sensitive patient data. We propose a framework integrating federated learning into distributed healthcare information systems (HIS) to enhance predictive analytics capabilities while preserving privacy. A comparative analysis of existing FL frameworks is presented alongside real-world use cases. Empirical evaluation demonstrates the balance achieved between model performance and data confidentiality.