JMIR Publications

ABSTRACT

The integrity and reliability of clinical research outcomes hinge on substantial amounts of data. However, their fragmented distribution across multiple contributors, coupled with ethical and regulatory difficulties, amplifies the challenges of accessing the most pertinent data. Despite federated learning’s potential to harness insights from fragmented data, its adoption is hindered by implementation complexities, scalability issues, and inclusivity challenges. To navigate these, this paper presents the Federated Learning For Everyone (FL4E) framework, a versatile and accessible ecosystem that promotes multi-stakeholder clinical research collaboration. FL4E simplifies the intricacies of federated learning with a user-friendly, modular structure that underpins an ecosystem-based design. An innovative ‘degree of federation’ feature strikes a balance between centralized and federated learning, paving the way for a tailored application in various healthcare settings. Through rigorous testing on real-world healthcare datasets, we have demonstrated the efficacy of FL4E. Our findings show that the degree of federation’s hybrid model achieves performance comparable to fully federated models without the associated significant overhead. Such results underscore the effectiveness of federation and suggest that FL4E has the potential to revolutionize collaborative clinical research by bridging the gap between centralized and federated approaches. The detailed implementation of FL4E, along with Docker configurations and related analyses, can be found on the corresponding GitHub page.