A Robust Pipeline for Differentially Private Federated Learning on Imbalanced Clinical Data using SMOTETomek and FedProx
Rodrigo Tertulino · Journal of the Brazilian Computer Society · 2026
Federated Learning (FL) offers a groundbreaking approach to collaborative health research, enabling model training on decentralized data while safeguarding patient privacy. FL offers formal security guarantees when combined with Differential Privacy (DP). The integration of these technologies, however, introduces a significant trade-off between privacy and clinical utility, a challenge further complicated by the severe class imbalance often present in medical datasets. The research presented herein addresses these interconnected issues through a systematic, multi-stage analysis. An FL framework was implemented for cardiovascular risk prediction, where initial experiments showed that standard methods struggled with imbalanced data, resulting in a Recall of zero. To overcome such a limitation, we first integrated the hybrid Synthetic Minority Over-sampling Technique with Tomek Links (SMOTETomek) at the client level, successfully developing a clinically useful model. Subsequently, the framework was optimized for non-IID data using a tuned FedProx algorithm. Our final results reveal a clear, non-linear trade-off between the privacy budget (ε) and model utility. An optimal operational region was identified where formal differential privacy guarantees with a single-digit privacy budget (ε ≈ 9.0 ) can be achieved while maintaining high clinical utility (Recall ≈ 77% ) and discriminative power (ROC-AUC ≈ 0.82 ). Although single-digit values ε are more relaxed than the tight budgets studied in theoretical DP literature, they represent a pragmatic operating point commonly reported in applied healthcare FL research. Ultimately, our study provides a practical methodological blueprint for creating effective, secure, and accurate diagnostic tools that can be applied to real-world, heterogeneous healthcare data.