Generating Clinically Realistic EHR Data via a Hierarchy- and Semantics-Guided Transformer

2025-02-28

AI Summary

Existing EHR generation methods treat medical codes as flat sequences, ignoring their inherent hierarchical structure and semantic descriptions, resulting in low clinical fidelity of synthetic data and suboptimal performance on downstream tasks. To address this, we propose HierGen—the first generative framework that jointly incorporates a clinical coding hierarchy graph and ClinicalBERT-derived semantic embeddings. HierGen employs a hierarchical graph neural network to model topological relationships among codes, integrates multi-source embeddings via a dedicated fusion mechanism, and leverages a Transformer-based decoder for interpretable, high-fidelity sequence generation. Evaluated on MIMIC-III and MIMIC-IV, HierGen significantly improves statistical distribution alignment with real EHRs, boosts chronic disease classification accuracy by 8.2%, and enables high-quality, privacy-preserving data augmentation.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Generating realistic synthetic electronic health records (EHRs) holds tremendous promise for accelerating healthcare research, facilitating AI model development and enhancing patient privacy. However, existing generative methods typically treat EHRs as flat sequences of discrete medical codes. This approach overlooks two critical aspects: the inherent hierarchical organization of clinical coding systems and the rich semantic context provided by code descriptions. Consequently, synthetic patient sequences often lack high clinical fidelity and have limited utility in downstream clinical tasks. In this paper, we propose the Hierarchy- and Semantics-Guided Transformer (HiSGT), a novel framework that leverages both hierarchical and semantic information for the generative process. HiSGT constructs a hierarchical graph to encode parent-child and sibling relationships among clinical codes and employs a graph neural network to derive hierarchy-aware embeddings. These are then fused with semantic embeddings extracted from a pre-trained clinical language model (e.g., ClinicalBERT), enabling the Transformer-based generator to more accurately model the nuanced clinical patterns inherent in real EHRs. Extensive experiments on the MIMIC-III and MIMIC-IV datasets demonstrate that HiSGT significantly improves the statistical alignment of synthetic data with real patient records, as well as supports robust downstream applications such as chronic disease classification. By addressing the limitations of conventional raw code-based generative models, HiSGT represents a significant step toward clinically high-fidelity synthetic data generation and a general paradigm suitable for interpretable medical code representation, offering valuable applications in data augmentation and privacy-preserving healthcare analytics.

Problem

• Generates realistic synthetic EHR data for healthcare research.
• Incorporates hierarchical and semantic information for clinical fidelity.
• Improves downstream applications like chronic disease classification.

Innovation

• Hierarchy- and Semantics-Guided Transformer (HiSGT) framework
• Graph neural network for hierarchy-aware embeddings
• Fusion with ClinicalBERT for semantic embeddings