PrivFusion: A Privacy-preserving Multi-Agent Framework for Harmonizing Distributed Datasets
For multi-site clinical machine learning, PrivFusion addresses the critical but overlooked bottleneck of data harmonization in a privacy-preserving manner.
PrivFusion automates harmonization of heterogeneous structured datasets across institutions before federated learning, reducing manual effort while effectively aligning multi-site COVID-19 data.
The growing availability of clinical data has increased the use of machine learning, yet centralized data aggregation is often infeasible for sensitive health information. Federated Learning (FL) offers a distributed alternative, but its adoption is limited by substantial heterogeneity across institutional datasets, making harmonization a critical but frequently overlooked prerequisite for multi-site analytics. We introduce PrivFusion, a privacy-preserving multi-agent framework that automates the harmonization of structured datasets prior to federated training. PrivFusion uses agents to analyze local data, cluster semantically similar features across sites, and provide iterative transformation recommendations until alignment is achieved. Evaluation across four heterogeneous COVID-19 datasets demonstrates that PrivFusion effectively and efficiently harmonizes multi-site data while substantially reducing manual effort.