FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting

Abstract Federated learning (FL) enables collaborative analysis of decentralized medical data while preserving patient privacy. However, the covariate shift from demographic and clinical differences can reduce model generalizability. We propose FedWeight, a novel FL framework that mitigates covariat...

Full description

Saved in:
Bibliographic Details
Main Authors: He Zhu, Jun Bai, Na Li, Xiaoxiao Li, Dianbo Liu, David L. Buckeridge, Yue Li
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01661-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849326863953952768
author He Zhu
Jun Bai
Na Li
Xiaoxiao Li
Dianbo Liu
David L. Buckeridge
Yue Li
author_facet He Zhu
Jun Bai
Na Li
Xiaoxiao Li
Dianbo Liu
David L. Buckeridge
Yue Li
author_sort He Zhu
collection DOAJ
description Abstract Federated learning (FL) enables collaborative analysis of decentralized medical data while preserving patient privacy. However, the covariate shift from demographic and clinical differences can reduce model generalizability. We propose FedWeight, a novel FL framework that mitigates covariate shift by reweighting patient data from the source sites using density estimators, allowing the trained model to better align with the distribution of the target site. To support unsupervised applications, we introduce FedWeight ETM, a federated embedded topic model. We evaluated FedWeight in cross-site FL on the eICU dataset and cross-dataset FL between eICU and MIMIC III. FedWeight consistently outperforms standard FL baselines in predicting ICU mortality, ventilator use, sepsis diagnosis, and length of stay. SHAP-based interpretation and ETM-based topic modeling reveal improved identification of clinically relevant characteristics and disease topics associated with ICU readmission.
format Article
id doaj-art-b17a38412bc644dfa7cf9ea06ac72fa8
institution Kabale University
issn 2398-6352
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-b17a38412bc644dfa7cf9ea06ac72fa82025-08-20T03:48:02ZengNature Portfolionpj Digital Medicine2398-63522025-05-018111910.1038/s41746-025-01661-8FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weightingHe Zhu0Jun Bai1Na Li2Xiaoxiao Li3Dianbo Liu4David L. Buckeridge5Yue Li6School of Computer Science, McGill UniversitySchool of Computer Science, McGill UniversityCommunity Health Sciences, Cumming School of Medicine, University of CalgaryElectrical and Computer Engineering, University of British ColumbiaSchool of Medicine, National University of SingaporeMila—Quebec AI InstituteSchool of Computer Science, McGill UniversityAbstract Federated learning (FL) enables collaborative analysis of decentralized medical data while preserving patient privacy. However, the covariate shift from demographic and clinical differences can reduce model generalizability. We propose FedWeight, a novel FL framework that mitigates covariate shift by reweighting patient data from the source sites using density estimators, allowing the trained model to better align with the distribution of the target site. To support unsupervised applications, we introduce FedWeight ETM, a federated embedded topic model. We evaluated FedWeight in cross-site FL on the eICU dataset and cross-dataset FL between eICU and MIMIC III. FedWeight consistently outperforms standard FL baselines in predicting ICU mortality, ventilator use, sepsis diagnosis, and length of stay. SHAP-based interpretation and ETM-based topic modeling reveal improved identification of clinically relevant characteristics and disease topics associated with ICU readmission.https://doi.org/10.1038/s41746-025-01661-8
spellingShingle He Zhu
Jun Bai
Na Li
Xiaoxiao Li
Dianbo Liu
David L. Buckeridge
Yue Li
FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
npj Digital Medicine
title FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
title_full FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
title_fullStr FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
title_full_unstemmed FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
title_short FedWeight: mitigating covariate shift of federated learning on electronic health records data through patients re-weighting
title_sort fedweight mitigating covariate shift of federated learning on electronic health records data through patients re weighting
url https://doi.org/10.1038/s41746-025-01661-8
work_keys_str_mv AT hezhu fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT junbai fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT nali fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT xiaoxiaoli fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT dianboliu fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT davidlbuckeridge fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting
AT yueli fedweightmitigatingcovariateshiftoffederatedlearningonelectronichealthrecordsdatathroughpatientsreweighting