Summary
Different hospitals collect health data that can help build smart tools to predict diseases, but privacy rules mean they can’t share patient details directly. The authors created a method that lets each hospital build its own model while working together by sharing only summary information. Their approach makes nearby hospitals’ models more similar because their patient data tends to be alike, which helps improve prediction accuracy. The method also focuses on optimizing a common measure called AUC, which tells how well the tool can distinguish sick and healthy patients. Tests with simulated and real data showed this teamwork helped improve the models, especially for hospitals close to each other.
federated learningAUC (Area Under Curve)personalized modelsprivacyhealthcare datagraph-based regularizationdata heterogeneityinfectious disease predictionmodel optimization
Authors
Mayu Hiraishi, Kensuke Tanioka, Toshio Shimokawa
Abstract
Accurate diagnostic and risk-prediction models are important for supporting clinical decision-making during infectious disease outbreaks. However, privacy and governance requirements may restrict patient-level data sharing across healthcare institutions, and data distributions often vary. Moreover, AUC is widely used to evaluate discriminative performance, motivating its direct optimization in model development. We propose geographically regularized AUC-maximizing personalized federated learning (GrAUC-PFL), which directly optimizes a smooth pairwise AUC surrogate to learn personalized models while keeping patient-level data local and accounting for institutional heterogeneity. Graph-based regularization encourages geographically neighboring institutions to have similar coefficient vectors while retaining a personalized models. Simulations and a real-data application suggest improved discriminative performance, particularly when geographically neighboring institutions have similar data-generating characteristics.