A Comprehensive Review on Privacy-Preserving Federated Learning Frameworks for Healthcare IoT Using Blockchain and Edge Intelligence

Abstract

The fast growth of the healthcare Internet of Things (IoT) solutions resulted in the generation of vast amounts of confidential medical data, with an equally growing concern regarding data protection and data use, amongst other factors. This review article will cover the most cutting- edge developments in federated learning (FL) frameworks for privacy- preserving blockchain integration and edge computing in healthcare. At a theoretical level, it became clear that the current centralised machine-learning methods are associated with the risks of data-privacy loss, regulatory impediments, and resource-constraint costs. In this situation, federated learning emerges as a very promising new paradigm capable of accommodating decentralised model training without raw data sharing, thereby further ensuring data privacy and security. In recent years, several contributions have been made combining secure characteristics for data privacy during training and aggregation and well-known techniques such as Fully Homomorphic Encryption (FHE), Secure Multi-Party Computation (SMPC), and Differential Privacy (DP). Moreover, the paper also touches upon the possibilities of blockchain to create secure, transparent, and immutable updates to the models using methodologies like Proof of Authority (PoA) and Delegated Proof of Stake (DPoS). Moreover, directing edge computing into the design could further enhance systems’ scalability and minimise latency within the healthcare domain. The text addresses the issues like non- IID data distribution, class imbalance, communication overhead, and convergence problems in the federated setting. Emerging methodologies such as cross-domain federated learning, multitask learning, and adaptive aggregation strategies were discussed to enhance the model’s generalisation and efficiency levels. The overall goal of the paper is to propose and develop a combination of federated learning together with blockchain and privacy-preserving technologies to enhance secure, scalable, and efficient health IoT systems.

How to cite this article:
Taha M, Rai A K. A Comprehensive Review on Privacy-Preserving Federated Learning Frameworks for Healthcare IoT Using Blockchain and Edge Intelligence. J Adv Res Cloud Comp Virtu Web Appl 2026; 9(2): 1-10.