Adoption of Federated Learning for Healthcare Informatics: Emerging Applications and Future Directions

The smart healthcare system has improved the patients quality of life (QoL), where records are being analyzed remotely by distributed stakeholders. It requires a voluminous exchange data for disease prediction via open communication channel, i.e., Internet to train artificial intelligence (AI) models efficiently and effectively. nature channels puts privacy at high risk affects model training collected centralized servers. To overcome this, an emerging concept, federated learning (FL) is viable solution. performs client nodes aggregates their results global model. concept local preserves privacy, confidentiality, integrity patient’s which contributes effectively process. applicability FL in domain various advantages, but it not been explored its extent. existing surveys majorly focused on role diverse applications, there exists no detailed or comprehensive survey informatics (HI). We present relative comparison recent with proposed survey. strengthen increase QoL patients, we FL-based layered architecture along case study electronic health (FL-EHR). discuss models, statistical security challenges adoption medical setups. Thus, review presents useful insights both academia practitioners investigate application HI ecosystems.