The paper discusses the future of digital health through the lens of federated learning (FL), a machine learning (ML) paradigm that enables collaborative training without centralizing data. The authors highlight the challenges and benefits of FL in addressing privacy and data governance issues in healthcare, particularly in the context of medical data silos and regulatory constraints. They explore how FL can enable large-scale, institutional validation and novel research on rare diseases, while maintaining patient privacy and data control. The paper also reviews current applications of FL in digital health, such as electronic health records, medical imaging, and drug discovery, and outlines the technical considerations and challenges, including data heterogeneity, privacy and security, traceability, and system architecture. The authors conclude by emphasizing the potential of FL to enhance precision medicine and improve patient outcomes while respecting privacy and governance concerns.The paper discusses the future of digital health through the lens of federated learning (FL), a machine learning (ML) paradigm that enables collaborative training without centralizing data. The authors highlight the challenges and benefits of FL in addressing privacy and data governance issues in healthcare, particularly in the context of medical data silos and regulatory constraints. They explore how FL can enable large-scale, institutional validation and novel research on rare diseases, while maintaining patient privacy and data control. The paper also reviews current applications of FL in digital health, such as electronic health records, medical imaging, and drug discovery, and outlines the technical considerations and challenges, including data heterogeneity, privacy and security, traceability, and system architecture. The authors conclude by emphasizing the potential of FL to enhance precision medicine and improve patient outcomes while respecting privacy and governance concerns.