Multi-center Medical Data Mining with FL-Net - A One-stop Shop for Federated Learning
FL-Net is a federated clinical research framework integrating harmonization, disclosure control, and audited workflows, targeting 800,000+ patients across 10 hospitals in 9 countries.
The authors derive five requirements from the literature and find that none of 14 analyzed federated learning frameworks fully satisfies them. FL-Net combines modular data harmonization, data discovery, disclosure control, securely built versioned FL-Net-Tools, and containerized federated workflow execution in a persistent network. It was evaluated with cross-study patient discovery across MIMIC and US-130 and reproducible, audited workflows with up to 50 concurrent clients. Development continues within the EU dAIbetes and Microb-AI-ome projects, aiming to cover over 800,000 patients across 10 hospitals in 9 countries.