Federated learning improves precision medicine by integrating genomics and imaging, suggesting subgroup fairness across borders.
We present Federated Multi-Omics Transformers (FMOT), a framework for cross-border precision medicine that fuses genomics, transcriptomics, proteomics, and imaging under privacy-preserving federated learning. FMOT introduces a subgroup-fairness head with equalized-odds regularization and auditable training. Synthetic multi-country experiments show monotonic AUROC gains, shrinking EO gaps, and manageable latency across rounds. We provide a mathematical formulation, trade-off analysis, and a governance-ready pipeline.
No takes yet. Share an insight, caveat, or question.
Murali Krishna Pasupuleti (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: