Learning the Graphical Structure of Electronic Health Records with Graph\n Convolutional Transformer

2019/06/11 by Edward Choi, Zhen Xu, Choi, Edward +12 · 6 citations
Computer Science · Decision Sciences · #Advanced Graph Neural Networks #Data Quality and Management #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning in Healthcare #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1906.04716

openalex publication_date 2019/06/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Effective modeling of electronic health records (EHR) is rapidly becoming an\nimportant topic in both academia and industry. A recent study showed that using\nthe graphical structure underlying EHR data (e.g. relationship between\ndiagnoses and treatments) improves the performance of prediction tasks such as\nheart failure prediction. However, EHR data do not always contain complete\nstructure information. Moreover, when it comes to claims data, structure\ninformation is completely unavailable to begin with. Under such circumstances,\ncan we still do better than just treating EHR data as a flat-structured\nbag-of-features? In this paper, we study the possibility of jointly learning\nthe hidden structure of EHR while performing supervised prediction tasks on EHR\ndata. Specifically, we discuss that Transformer is a suitable basis model to\nlearn the hidden EHR structure, and propose Graph Convolutional Transformer,\nwhich uses data statistics to guide the structure learning process. The\nproposed model consistently outperformed previous approaches empirically, on\nboth synthetic data and publicly available EHR data, for various prediction\ntasks such as graph reconstruction and readmission prediction, indicating that\nit can serve as an effective general-purpose representation learning algorithm\nfor EHR data.\n

Citations

• Semi-Supervised Classification with Graph Convolutional Networks
• Clinical Intervention Prediction and Understanding using Deep Networks
• RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism

Cited by

• Patient-level Information Extraction by Consistent Integration of Textual and Tabular Evidence with Bayesian Networks
• CURENet: Combining Unified Representations for Efficient Chronic Disease Prediction
• KAT-GNN: A Knowledge-Augmented Temporal Graph Neural Network for Risk Prediction in Electronic Health Records
• Multi-Ontology Integration with Dual-Axis Propagation for Medical Concept Representation
• Structure-aware Hypergraph Transformer for Diagnosis Prediction in Electronic Health Records
• Intellectual Property in Graph-Based Machine Learning as a Service: Attacks and Defenses

Related

• Structure-aware Hypergraph Transformer for Diagnosis Prediction in Electronic Health Records
• Medical Applications of Graph Convolutional Networks Using Electronic Health Records: A Survey
• Predictive Modeling with Temporal Graphical Representation on Electronic Health Records
• Representation Learning of EHR Data via Graph-Based Medical Entity Embedding
• Multi-task Heterogeneous Graph Learning on Electronic Health Records