The era of healthcare system in the next generation is Healthcare 5.0. This approach uses latest technologies of the Artificial Intelligence, Internet of Medical Things (IoMT), robotics, and cloud computing technologies in the view of human centered care. Traditional system failure to cover data interoperability and to share data effectively with varied terminologies and data formats. Using Federated Learning (FL) which allows training on decentralized data without sharing raw data, though it faces issues with non-IID datasets common in healthcare. To address these challenges, the chapter suggests combining ontology based data harmonization with Federated Transfer Learning (FTL). Ontologies help standardize diverse data for better exchange, while FTL allows models to adapt to local data characteristics. This hybrid approach enables collaboration between healthcare providers while protecting patient privacy and enhancing model accuracy, leading to improved healthcare outcomes. The chapter also discusses the theoretical and practical aspects of implementing this strategy. Experimental results show that the proposed hybrid framework attains a medium performance improvement up to 23.6% in predictive accuracy compared with traditional federated learning methods in non-IID environment. This adaptable and privacy conscious model encourages collaboration across institutions, facilitating the development of sophisticated, patient focused healthcare intelligence in the era of Healthcare 5.0.