TAMER: A Test-Time Adaptive MoE-Driven Framework for EHR Representation Learning

TL;DR

TAMER is a novel framework that combines a Mixture-of-Experts architecture with test-time adaptation to improve real-time, personalized predictions from Electronic Health Records by addressing patient heterogeneity and distribution shifts.

Contribution

TAMER introduces a co-designed MoE and TTA framework specifically for EHR representation learning, enabling dynamic adaptation to evolving patient data.

Findings

Consistently improves predictive accuracy across multiple EHR datasets.

Enhances mortality and readmission risk prediction performance.

Effectively handles patient heterogeneity and distribution shifts.

Abstract

We propose TAMER, a Test-time Adaptive MoE-driven framework for Electronic Health Record (EHR) Representation learning. TAMER introduces a framework where a Mixture-of-Experts (MoE) architecture is co-designed with Test-Time Adaptation (TTA) to jointly mitigate the intertwined challenges of patient heterogeneity and distribution shifts in EHR modeling. The MoE focuses on latent patient subgroups through domain-aware expert specialization, while TTA enables real-time adaptation to evolving health status distributions when new patient samples are introduced. Extensive experiments across four real-world EHR datasets demonstrate that TAMER consistently improves predictive performance for both mortality and readmission risk tasks when combined with diverse EHR modeling backbones. TAMER offers a promising approach for dynamic and personalized EHR-based predictions in practical clinical settings.