Advances in machine learning (ML), deep-learning (DL) algorithms, the Internet of Things (IoT), and artificial intelligence (AI) have enabled the development of reliable and scalable diagnostic and predictive models based on medical data. This has led to significant improvements in healthcare quality. However, adopting AI-based medical applications still faces challenges such as security, privacy, and quality of service standards. Several ML techniques have been incorporated into healthcare systems to develop stable data processing methods. However, applying ML to precise clinical data is challenging due to the lack of precise clinical data. Protecting patient privacy and data confidentiality is the main drawback of electronic medical record transmission. Federated learning (FL) offers a promising solution by allowing multiple clients to work together without transferring data to a single location. FL can lower communication and storage costs while maintaining high user privacy. Recent breakthroughs in FL have shown numerous success stories for smart healthcare applications that use data-driven insights to improve clinical care quality. The most recent innovations in intelligent healthcare include FL, AI, and Explainable Artificial Intelligence (XAI), which have the potential to reduce various limits and issues in the healthcare system. This chapter discusses the current and future state of FL technology in medical applications, research trends, and findings that illustrate the complexity of creating reliable and scalable FL models. However, obstacles remain in deploying a true FL system in IoT networks. Research is paving the way for breakthroughs in various healthcare domains to demonstrate that prototypes trained with FL methods can reach dependable performance, minimizing the impact on security and privacy. Decentralized networks are becoming increasingly important in today's healthcare business due to effective privacy issues.