aiqbench analysis
This study demonstrates that federated training of tokenized generative event models effectively overcomes data silos in electronic health records while maintaining high predictive performance across independent hospital systems. The research highlights a critical technical finding where these transportable representations significantly outperform conventional supervised models when transferring knowledge between sites with heterogeneous datasets. While the approach is technically feasible, it identifies learning robust, generalizable representations as the primary remaining challenge for scaling AI applications in fragmented healthcare environments.
Federated generative event models for tokenized electronic health records
arXiv:2608.02939v1 Announce Type: new Abstract: Electronic health record foundation models are limited by institutionally siloed data and substantial performance degradation under cross-site transfer. We evaluated federated training of tokenized generative event models (GEMs) across 122,251 intensive care hospitalizations from three independent health systems harmonized to the Common Longitudinal ICU Data Format. Models were assessed on 12 post-24-hour clinical prediction tasks using within-site, cross-site, centralized, and federated training configurations. GEMs achieved the highest mean within-site and cross-site ROC-AUC and were substantially more transportable than conventional supervised models: their average cross-site penalties were 0.025 ROC-AUC and 0.027 PR-AUC, compared with 0.079 and 0.089 for LightGBM. Federated Learning (FedAvg and FedAvgM) approached the performance of centralized GEM training, with most gains obtained within 5-10 communication rounds. However, centralized multi-site training provided only modest improvements over complete local training. Multi-site models were most useful when local training data were limited, with their advantage narrowing as institutional data accumulated. These findings show that federated GEM training is technically feasible and preserves most centralized performance, but that the main open challenge is learning transportable representations to translate larger, but heterogeneous data from multiple health systems into a reliable target-site benefit.
Authors:Michael C. Burkhart, Luke Solo, Inhyeok Lee, S'Khaja Charles, Zewei "Whiskey" Liao, Kaveri Chhikara, Dema Therese, Wan-Ting Liao, Catherine A. Gao, William F. Parker, Brett K. Beaulieu-Jones
Abstract:Electronic health record foundation models are limited by institutionally siloed data and substantial performance degradation under cross-site transfer. We evaluated federated training of tokenized generative event models (GEMs) across 122,251 intensive care hospitalizations from three independent health systems harmonized to the Common Longitudinal ICU Data Format. Models were assessed on 12 post-24-hour clinical prediction tasks using within-site, cross-site, centralized, and federated training configurations. GEMs achieved the highest mean within-site and cross-site ROC-AUC and were substantially more transportable than conventional supervised models: their average cross-site penalties were 0.025 ROC-AUC and 0.027 PR-AUC, compared with 0.079 and 0.089 for LightGBM. Federated Learning (FedAvg and FedAvgM) approached the performance of centralized GEM training, with most gains obtained within 5-10 communication rounds. However, centralized multi-site training provided only modest improvements over complete local training. Multi-site models were most useful when local training data were limited, with their advantage narrowing as institutional data accumulated. These findings show that federated GEM training is technically feasible and preserves most centralized performance, but that the main open challenge is learning transportable representations to translate larger, but heterogeneous data from multiple health systems into a reliable target-site benefit.
Machine Learning (cs.LG); Computers and Society (cs.CY)
arXiv:2608.02939 [cs.LG]
(or arXiv:2608.02939v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.02939 arXiv-issued DOI via DataCite (pending registration)
From: Michael C. Burkhart [view email] [v1] Mon, 3 Aug 2026 22:54:31 UTC (18,474 KB)
Full text under the publisher's open licence · 253 words
Related coverage
Digital States 2026: AI Moves From Experiment to Infrastructure
govtech.com · Sep 24, 2026
INTERVenE: Temporal-Abstraction-Interval Based Transformers for Short-Horizon Medical Event Prediction
arXiv · Sep 1, 2026
Industrial Tokenization for LLM-Based Health Intelligence: A Federated Architecture for Industrial Evidence Integration
arXiv · Jul 28, 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · Aug 27, 2026