If you don't see an answer to your question, send us an email from our contact section.
Yes, Fed-BioMed is a fully open-source project, initiated by Inria and Université Côte d’Azur. You can reach the project’s GitHub page from here.
Fed-BioMed is distributed under the Apache-2.0 License.
The software is originally developed by Inria and Université Côte d’Azur, with contributions from the Fed-BioMed community.
Fed-BioMed is designed for collaborative health data programs. It enables hospitals, biobanks, registries, clinical networks, and any organization holding sensitive health or subject data to collaborate with research teams and industry partners in a privacy-preserving way - without sharing or moving that data..
Fed-BioMed is a framework developed in Python. It is used for the deployment and governance of federated workflows between the researcher and the medical data owner, such as federated learning and federated analytics.
Fed-BioMed is available as an open-source project. You can install it from the official pip package or build it from the source code available on GitHub. The documentation provides installation instructions for both pip and source code, as well as on installing different Fed-BioMed components, such as the node and researcher.
Fed-BioMed is designed to support privacy-preserving and governance-aware federated learning in contexts where GDPR compliance is important. However, GDPR compliance does not depend on the software alone. It also depends on how the project is deployed, the legal basis for medical data processing, the security measures in place, and the responsibilities of the participating institutions. Users should therefore assess GDPR compliance within their own legal, technical, and organizational context.
Yes, Fed-BioMed is designed with real-world medical research applications in mind. However, using it with real medical data requires appropriate legal approvals. Each participating institution should ensure that its data use complies with the respective regulations and local policies.
Yes. Fed-BioMed can be used for federated learning experiments, as well as federated analytics and data discovery. Researchers are able to customize the nodes and data included in the federated training to simulate different scenarios.
Fed-BioMed is mainly intended for collaborative research projects involving distributed medical data. This allows clinicians, data scientists, or people from any other background to develop and run federated learning or federated analytics experiments. Developers and engineers are also very welcome to contribute to the development of Fed-BioMed.
You can contact the Fed-BioMed team through the official communication channels provided on the Support & Contact section, or GitHub repository. The team may be able to provide guidance, discuss collaborations, or help evaluate whether Fed-BioMed is suitable for your research use case.
If you use Fed-BioMed in your research, please kindly refer to our page About > How to cite
Mostly yes. Fed-BioMed is developed to meet medical use cases requirements. While in principle it could be used for other domains where data cannot easily be centralized and federated learning is needed, it does not intend to be a general usage federated learning framework.
No. Fed-BioMed is not a SaaS platform. It is an open-source framework that can be deployed by institutions or research teams in their own environments.
Fed-BioMed requires Python and the relevant machine learning packages. For real-world deployments the version packaged in Docker containers with VPN is recommended. Fed-BioMed is developed and tested mainly on Linux and macOS, and can be used on Windows through WSL.
The deployment effort depends on the complexity of the project. A local or experimental setup is quick to install, while a real multi-institution medical deployment may require additional work, especially on institutional approvals.
Fed-BioMed focuses on collaborative health data programs for sensitive real-world environments. Compared with general-purpose federated learning frameworks, it places strong emphasis on medical use cases, distributed data governance, security, deployment constraints, and usability for research collaborations.
No. Fed-BioMed is an open-source project for research, and is not a commercial product. For specific collaborations, support, integration, or research partnerships, users can contact the Fed-BioMed team directly to discuss what may be possible.