Recent technological advancements have considerately improved healthcare systems to provide various intelligent healthcare services and improve the quality of life. Federated learning (FL), a new branch of artificial intelligence (AI), opens opportunities to deal with privacy issues in healthcare systems and exploit data and computing resources available at distributed devices. Additionally, the Metaverse, through integrating emerging technologies, such as AI, cloud edge computing, Internet of Things (IoT), blockchain, and semantic communications, has transformed many vertical domains in general and the healthcare sector in particular. Obviously, FL shows many benefits and provides new opportunities for conventional and Metaverse healthcare, motivating us to provide a survey on the usage of FL for Metaverse healthcare systems. First, we present preliminaries to IoT-based healthcare systems, FL in conventional healthcare, and Metaverse healthcare. The benefits of FL in Metaverse healthcare are then discussed, from improved privacy and scalability, better interoperability, better data management, and extra security to automation and low-latency healthcare services. Subsequently, we discuss several applications pertaining to FL-enabled Metaverse healthcare, including medical diagnosis, patient monitoring, medical education, infectious disease, and drug discovery. Finally, we highlight significant challenges and potential solutions toward the realization of FL in Metaverse healthcare.

翻译：近年来的技术进步显著提升了医疗系统，使其能够提供多样化的智能医疗服务并改善生活质量。作为人工智能的一个新兴分支，联邦学习为应对医疗系统中的隐私问题、利用分布式设备上的数据和计算资源提供了新的机遇。此外，元宇宙通过整合人工智能、云边计算、物联网、区块链和语义通信等新兴技术，已广泛变革了包括医疗在内的多个垂直领域。显然，联邦学习在传统医疗和元宇宙医疗中均展现出诸多优势并带来新的机遇，这促使我们对其在元宇宙医疗系统中的应用进行综述。首先，我们介绍了基于物联网的医疗系统、传统医疗中的联邦学习以及元宇宙医疗的基础知识。随后讨论了联邦学习在元宇宙医疗中的优势，包括提升的隐私性和可扩展性、更优的互操作性、更高效的数据管理、增强的安全性，以及自动化与低延迟医疗服务。接着，我们探讨了联邦学习赋能元宇宙医疗的若干应用场景，涵盖医疗诊断、患者监测、医学教育、传染病防控和药物发现。最后，我们指出了在元宇宙医疗中实现联邦学习所面临的重要挑战及潜在解决方案。