From EHRs to Patient Pathways: Scalable Modeling of Longitudinal Health Trajectories with LLMs

2025-06-05

AI Summary

Integrating heterogeneous, multi-source electronic health records (EHR) to model long-term patient health trajectories remains a major challenge in healthcare systems. This paper introduces EHR2Path, the first framework that transforms raw EHR data into structured, interpretable patient pathway representations, enabling long-horizon modeling and multi-step longitudinal simulation. Its key contributions are: (1) a topic-aware, long-term temporal summarization mechanism that captures global contextual dependencies with high token efficiency; and (2) a unified architecture integrating large language model–based sequence modeling, temporal structural encoding, topic-driven dynamic summary token generation, and multimodal EHR alignment. Evaluated on next-time-step prediction and multi-week longitudinal simulation tasks, EHR2Path consistently outperforms state-of-the-art baselines across diverse clinical outcomes—including vital signs, laboratory test results, and length of hospital stay—with up to a 12.3% improvement in AUC.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Healthcare systems face significant challenges in managing and interpreting vast, heterogeneous patient data for personalized care. Existing approaches often focus on narrow use cases with a limited feature space, overlooking the complex, longitudinal interactions needed for a holistic understanding of patient health. In this work, we propose a novel approach to patient pathway modeling by transforming diverse electronic health record (EHR) data into a structured representation and designing a holistic pathway prediction model, EHR2Path, optimized to predict future health trajectories. Further, we introduce a novel summary mechanism that embeds long-term temporal context into topic-specific summary tokens, improving performance over text-only models, while being much more token-efficient. EHR2Path demonstrates strong performance in both next time-step prediction and longitudinal simulation, outperforming competitive baselines. It enables detailed simulations of patient trajectories, inherently targeting diverse evaluation tasks, such as forecasting vital signs, lab test results, or length-of-stay, opening a path towards predictive and personalized healthcare.

Problem

• Modeling longitudinal health trajectories from diverse EHR data
• Predicting future patient pathways with structured representations
• Improving efficiency and accuracy in health trajectory simulations

Innovation

• Transforms EHR data into structured representations
• Uses topic-specific summary tokens for efficiency
• Predicts future health trajectories holistically