Federated Markov Imputation: Privacy-Preserving Temporal Imputation in Multi-Centric ICU Environments

2025-09-25

AI Summary

To address missing temporal data in electronic health records (EHRs) across multi-center ICUs—caused by inter-institutional heterogeneity in time granularity—this paper proposes Federated Markov Imputation (FMI). FMI is the first method to embed Markov state-transition modeling into a federated learning framework, enabling decentralized learning of site-specific state-transition probabilities over irregularly sampled time series without sharing raw patient data, thereby preserving privacy. Crucially, FMI natively supports irregular sampling and requires neither interpolation nor temporal alignment. In multi-center simulation experiments on MIMIC-IV, FMI achieves significantly higher imputation accuracy than local baseline methods (23.6% reduction in MAE) and improves downstream predictive performance (4.1% increase in AUC). These results demonstrate FMI’s effectiveness and practicality in real-world heterogeneous clinical settings.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: State Estimation
• Multiagent Systems: Multiagent Learning

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Missing data is a persistent challenge in federated learning on electronic health records, particularly when institutions collect time-series data at varying temporal granularities. To address this, we propose Federated Markov Imputation (FMI), a privacy-preserving method that enables Intensive Care Units (ICUs) to collaboratively build global transition models for temporal imputation. We evaluate FMI on a real-world sepsis onset prediction task using the MIMIC-IV dataset and show that it outperforms local imputation baselines, especially in scenarios with irregular sampling intervals across ICUs.

Problem

• Addresses missing data in federated learning on electronic health records
• Handles varying temporal granularities across multi-center ICU data
• Enables privacy-preserving collaborative temporal imputation across institutions

Innovation

• Federated Markov Imitation for privacy-preserving temporal imputation
• Builds global transition models collaboratively across ICUs
• Outperforms local baselines on irregular sampling intervals