aiqbench analysis
This study rigorously benchmarks five Federated Learning strategies for predicting in-hospital mortality using the MIMIC-IV dataset under realistic non-IID and imbalanced clinical conditions. While FedProx emerges as the most robust federated approach regarding AUC metrics, centralized training still retains a slight predictive edge despite privacy constraints. The analysis highlights significant performance disparities across different care units, underscoring that no single algorithm universally dominates when dealing with heterogeneous medical data distributions.
A Comparative Benchmark of Federated Learning Strategies for Mortality Prediction on Heterogeneous and Imbalanced Clinical Data
arXiv:2509.10517v3 Announce Type: replace-cross Abstract: Machine learning can predict in-hospital mortality, but data privacy and the statistical heterogeneity of clinical data hamper its use. Federated Learning (FL) is privacy-preserving, yet its behavior under non-IID and imbalanced conditions needs scrutiny. We benchmark five FL strategies - FedAvg, FedProx, FedAdagrad, FedAdam, and FedCluster - for mortality prediction on the MIMIC-IV dataset, partitioning 466,351 admissions across five care units to induce a realistic non-IID setting and enriching the features with an 11-item first-24-hour laboratory panel. At a prevalence of 1.98%, we adopt AUC-ROC and AUC-PR as primary, threshold-independent metrics rather than F1. Over 50 rounds and five random seeds, FedProx attains the best AUC-ROC (0.897) and mean AUC-PR (0.230), with paired t-tests confirming its AUC-ROC lead is significant against every other strategy; on F1, however, FedCluster (0.280) narrowly surpasses FedProx (0.273), so no single strategy dominates every metric. The best centralized baseline achieves AUC-ROC and AUC-PR of 0.929 and 0.312, respectively, significantly outperforming FedProx on both. A per-client breakdown shows the global model does not serve all care units equally (AUC-ROC 0.809-0.902), with the smallest, most clinically distinct clients faring worst. We conclude that regularization-based methods such as FedProx are the more robust federated choice, while centralization retains a slight predictive edge, and that per-client heterogeneity and evaluation-set construction deserve attention independent of aggregate numbers.
Abstract:Machine learning can predict in-hospital mortality, but data privacy and the statistical heterogeneity of clinical data hamper its use. Federated Learning (FL) is privacy-preserving, yet its behavior under non-IID and imbalanced conditions needs scrutiny. We benchmark five FL strategies - FedAvg, FedProx, FedAdagrad, FedAdam, and FedCluster - for mortality prediction on the MIMIC-IV dataset, partitioning 466,351 admissions across five care units to induce a realistic non-IID setting and enriching the features with an 11-item first-24-hour laboratory panel. At a prevalence of 1.98%, we adopt AUC-ROC and AUC-PR as primary, threshold-independent metrics rather than F1. Over 50 rounds and five random seeds, FedProx attains the best AUC-ROC (0.897) and mean AUC-PR (0.230), with paired t-tests confirming its AUC-ROC lead is significant against every other strategy; on F1, however, FedCluster (0.280) narrowly surpasses FedProx (0.273), so no single strategy dominates every metric. The best centralized baseline achieves AUC-ROC and AUC-PR of 0.929 and 0.312, respectively, significantly outperforming FedProx on both. A per-client breakdown shows the global model does not serve all care units equally (AUC-ROC 0.809-0.902), with the smallest, most clinically distinct clients faring worst. We conclude that regularization-based methods such as FedProx are the more robust federated choice, while centralization retains a slight predictive edge, and that per-client heterogeneity and evaluation-set construction deserve attention independent of aggregate numbers.
16 pages, 4 figures, 9 tables. Substantially revised: a patient-level split removes data leakage; AUC-ROC/AUC-PR are the primary metrics; and all experiments were rerun across 5 random seeds with paired t-tests. Code: this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
I.2.6; I.2.11; J.3
arXiv:2509.10517 [cs.LG]
(or arXiv:2509.10517v3 [cs.LG] for this version)
Full text under the publisher's open licence · 273 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
When Is Shallow Enough? Adaptive Split Federated Learning with Client-Specific Sufficiency Estimation
arXiv · Aug 19, 2026
News Archive 2026
Quantum Computing Report · Jul 25, 2026
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · Sep 17, 2026