Readable research linked to original sources

Articles

Browse normalized, publication-ready research with direct links to its evidence and source.

2 articles

Newest first

Research Article 2026-03-08CC BYFree full text

Distribution-Aware Federated Learning for Diabetes Prediction Using Tabular Clinical Data Under Non-IID and Class-Imbalanced Settings

Rohul Amin, Md. Mehedi Hasan Rana, Sumya Aktar

Federated learning (FL) enables collaborative clinical model training without centralized data sharing, yet its deployment is hindered by statistical heterogeneity (non-IID data) and inherent class imbalance across healthcare institutions. Conventional aggregation strategies such as FedAvg and FedProx weight client updates solely by dataset size, ignoring class distributions and thereby biasing the global model to…

bioRxiv bioRxiv

Open article record in new tab ↗

Research Article 2025-08-06CC BYFree full text

Federated Learning for ICU Mortality Prediction: Balancing Accuracy and Privacy in a Multi-Hospital Setting

Yassir Benhammou, Suman Kalyan, Sujay Kumar

Intensive care units (ICUs) manage critically ill patients whose clinical outcomes rely on timely and accurate decision-making. Predictive modeling using electronic health records (EHRs) has shown promise in forecasting adverse events such as in-hospital mortality. However, constructing robust and generalizable models typically requires large-scale, multi-institutional datasets. Privacy regulations such as HIPAA a…

bioRxiv bioRxiv

Open article record in new tab ↗