An embedding-based federated learning system with a frozen haematology foundation model and personalised aggregation (FedMAP) improves iron deficiency prediction across two non-IID clinical sites over local-only training.
Red cell indices
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Embedding-Based Federated Learning with Runtime Governance for Iron Deficiency Prediction
An embedding-based federated learning system with a frozen haematology foundation model and personalised aggregation (FedMAP) improves iron deficiency prediction across two non-IID clinical sites over local-only training.