DT-BEHRT: Disease Trajectory-aware Transformer for Interpretable Patient Representation Learning

2026-03-10

AI Summary

Existing approaches to modeling electronic health records struggle to capture the heterogeneity of medical codes across diverse clinical contexts, often overlooking diagnosis-centered interactions within organ systems and asynchronous disease progression patterns. To address these limitations, this work proposes DT-BEHRT, a novel architecture that integrates graph-based and sequential modeling by decoupling disease trajectories, incorporating an organ system–aware mechanism for diagnosis interactions, and leveraging trajectory-level masking alongside medical ontology–guided ancestor prediction during pretraining. The proposed method achieves superior predictive performance across multiple benchmarks, yielding patient representations that exhibit strong semantic coherence and clinical interpretability, aligning closely with physicians’ disease-centered clinical reasoning.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The growing adoption of electronic health record (EHR) systems has provided unprecedented opportunities for predictive modeling to guide clinical decision making. Structured EHRs contain longitudinal observations of patients across hospital visits, where each visit is represented by a set of medical codes. While sequence-based, graph-based, and graph-enhanced sequence approaches have been developed to capture rich code interactions over time or within the same visits, they often overlook the inherent heterogeneous roles of medical codes arising from distinct clinical characteristics and contexts. To this end, in this study we propose the Disease Trajectory-aware Transformer for EHR (DT-BEHRT), a graph-enhanced sequential architecture that disentangles disease trajectories by explicitly modeling diagnosis-centric interactions within organ systems and capturing asynchronous progression patterns. To further enhance the representation robustness, we design a tailored pre-training methodology that combines trajectory-level code masking with ontology-informed ancestor prediction, promoting semantic alignment across multiple modeling modules. Extensive experiments on multiple benchmark datasets demonstrate that DT-BEHRT achieves strong predictive performance and provides interpretable patient representations that align with clinicians'disease-centered reasoning. The source code is publicly accessible at https://github.com/GatorAIM/DT-BEHRT.git.

Problem

• Electronic Health Records
• Patient Representation Learning
• Disease Trajectory
• Medical Code Heterogeneity
• Interpretable Modeling

Innovation

• Disease Trajectory
• Transformer
• Graph-enhanced Sequential Model
• Ontology-informed Pre-training
• Interpretable Representation