A review of recent advances in privacy-preserving federated learning for healthcare applications

Abstract

The rapid adoption of artificial intelligence (AI) in healthcare, driven by the increasing availability of electronic health records (EHRs), medical imaging systems, genomic data, and wearable devices, has created significant challenges related to patient privacy, data security, and regulatory compliance. Traditional centralised machine learning (ML) approaches often require large-scale data sharing, which increases the risk of sensitive medical information exposure. Federated learning (FL) has emerged as a promising distributed learning paradigm that enables collaborative model training across multiple institutions without transferring raw patient data. Recent advances in privacy-preserving FL have further strengthened its applicability in healthcare through the integration of techniques such as differential privacy (DP), homomorphic encryption (HE), secure aggregation, blockchain-assisted frameworks, and secure multi-party computation. This study critically reviews recent developments in privacy-preserving FL for healthcare applications published between 2024 and early 2026, with particular emphasis on medical imaging, EHRs analytics, genomics, and wearable health monitoring systems. In addition, major technical and security challenges, including non-independent and identically distributed (non-IID) data distributions, communication overhead, scalability issues, gradient leakage, model poisoning, and membership inference attacks, are comprehensively analysed. Regulatory, ethical, and interoperability concerns associated with cross-institutional collaboration are also discussed. Finally, current research gaps, emerging trends, and future directions for developing secure, scalable, and trustworthy federated healthcare systems are identified.