Patient-level Information Extraction by Consistent Integration of Textual and Tabular Evidence with Bayesian Networks

TL;DR

This paper introduces a Bayesian network-based method for integrating structured tabular data and unstructured clinical notes in electronic health records to improve patient information extraction and model interpretability.

Contribution

It presents a novel probabilistic fusion approach using virtual evidence and a consistency node to enhance the calibration and integration of multi-modal EHR data.

Findings

Improved prediction calibration with the consistency node.

Effective integration of textual and tabular data in EHRs.

Demonstrated potential on the SimSUM benchmark dataset.

Abstract

Electronic health records (EHRs) form an invaluable resource for training clinical decision support systems. To leverage the potential of such systems in high-risk applications, we need large, structured tabular datasets on which we can build transparent feature-based models. While part of the EHR already contains structured information (e.g. diagnosis codes, medications, and lab results), much of the information is contained within unstructured text (e.g. discharge summaries and nursing notes). In this work, we propose a method for multi-modal patient-level information extraction that leverages both the tabular features available in the patient's EHR (using an expert-informed Bayesian network) as well as clinical notes describing the patient's symptoms (using neural text classifiers). We propose the use of virtual evidence augmented with a consistency node to provide an interpretable, probabilistic fusion of the models' predictions. The consistency node improves the calibration of the final predictions compared to virtual evidence alone, allowing the Bayesian network to better adjust the neural classifier's output to handle missing information and resolve contradictions between the tabular and text data. We show the potential of our method on the SimSUM dataset, a simulated benchmark linking tabular EHRs with clinical notes through expert knowledge.