skip to main content


The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 5:00 PM ET until 11:00 PM ET on Friday, June 21 due to maintenance. We apologize for the inconvenience.

Title: Embedding electronic health records onto a knowledge network recognizes prodromal features of multiple sclerosis and predicts diagnosis
Abstract Objective

Early identification of chronic diseases is a pillar of precision medicine as it can lead to improved outcomes, reduction of disease burden, and lower healthcare costs. Predictions of a patient’s health trajectory have been improved through the application of machine learning approaches to electronic health records (EHRs). However, these methods have traditionally relied on “black box” algorithms that can process large amounts of data but are unable to incorporate domain knowledge, thus limiting their predictive and explanatory power. Here, we present a method for incorporating domain knowledge into clinical classifications by embedding individual patient data into a biomedical knowledge graph.

Materials and Methods

A modified version of the Page rank algorithm was implemented to embed millions of deidentified EHRs into a biomedical knowledge graph (SPOKE). This resulted in high-dimensional, knowledge-guided patient health signatures (ie, SPOKEsigs) that were subsequently used as features in a random forest environment to classify patients at risk of developing a chronic disease.


Our model predicted disease status of 5752 subjects 3 years before being diagnosed with multiple sclerosis (MS) (AUC = 0.83). SPOKEsigs outperformed predictions using EHRs alone, and the biological drivers of the classifiers provided insight into the underpinnings of prodromal MS.


Using data from EHR as input, SPOKEsigs describe patients at both the clinical and biological levels. We provide a clinical use case for detecting MS up to 5 years prior to their documented diagnosis in the clinic and illustrate the biological features that distinguish the prodromal MS state.

more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ;
Publisher / Repository:
Oxford Academic
Date Published:
Journal Name:
Journal of the American Medical Informatics Association
Page Range / eLocation ID:
424 to 434
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    In electronic health records (EHRs) data analysis, nonparametric regression and classification using International Classification of Disease (ICD) codes as covariates remain understudied. Automated methods have been developed over the years for predicting biomedical responses using EHRs, but relatively less attention has been paid to developing patient similarity measures that use ICD codes and chronic conditions, where a chronic condition is defined as a set of ICD codes. We address this problem by first developing a string kernel function for measuring the similarity between a pair of primary chronic conditions, represented as subsets of ICD codes. Second, we extend this similarity measure to a family of covariance functions on subsets of chronic conditions. This family is used in developing Gaussian process (GP) priors for Bayesian nonparametric regression and classification using diagnoses and other demographic information as covariates. Markov chain Monte Carlo (MCMC) algorithms are used for posterior inference and predictions. The proposed methods are tuning free, so they are ideal for automated prediction of biomedical responses depending on chronic conditions. We evaluate the practical performance of our method on EHR data collected from 1660 patients at the University of Iowa Hospitals and Clinics (UIHC) with six different primary cancer sites. Our method provides better sensitivity and specificity than its competitors in classifying different primary cancer sites and estimates the marginal associations between chronic conditions and primary cancer sites.

    more » « less
  2. Abstract

    Timely and accurate referral of end-stage heart failure patients for advanced therapies, including heart transplants and mechanical circulatory support, plays an important role in improving patient outcomes and saving costs. However, the decision-making process is complex, nuanced, and time-consuming, requiring cardiologists with specialized expertise and training in heart failure and transplantation.

    In this study, we propose two logistic tensor regression-based models to predict patients with heart failure warranting evaluation for advanced heart failure therapies using irregularly spaced sequential electronic health records at the population and individual levels. The clinical features were collected at the previous visit and the predictions were made at the very beginning of the subsequent visit. Patient-wise ten-fold cross-validation experiments were performed. Standard LTR achieved an average F1 score of 0.708, AUC of 0.903, and AUPRC of 0.836. Personalized LTR obtained an F1 score of 0.670, an AUC of 0.869 and an AUPRC of 0.839. The two models not only outperformed all other machine learning models to which they were compared but also improved the performance and robustness of the other models via weight transfer. The AUPRC scores of support vector machine, random forest, and Naive Bayes are improved by 8.87%, 7.24%, and 11.38%, respectively.

    The two models can evaluate the importance of clinical features associated with advanced therapy referral. The five most important medical codes, including chronic kidney disease, hypotension, pulmonary heart disease, mitral regurgitation, and atherosclerotic heart disease, were reviewed and validated with literature and by heart failure cardiologists. Our proposed models effectively utilize EHRs for potential advanced therapies necessity in heart failure patients while explaining the importance of comorbidities and other clinical events. The information learned from trained model training could offer further insight into risk factors contributing to the progression of heart failure at both the population and individual levels.

    more » « less
  3. Abstract INTRODUCTION

    Identifying mild cognitive impairment (MCI) patients at risk for dementia could facilitate early interventions. Using electronic health records (EHRs), we developed a model to predict MCI to all‐cause dementia (ACD) conversion at 5 years.


    Cox proportional hazards model was used to identify predictors of ACD conversion from EHR data in veterans with MCI. Model performance (area under the receiver operating characteristic curve [AUC] and Brier score) was evaluated on a held‐out data subset.


    Of 59,782 MCI patients, 15,420 (25.8%) converted to ACD. The model had good discriminative performance (AUC 0.73 [95% confidence interval (CI) 0.72–0.74]), and calibration (Brier score 0.18 [95% CI 0.17–0.18]). Age, stroke, cerebrovascular disease, myocardial infarction, hypertension, and diabetes were risk factors, while body mass index, alcohol abuse, and sleep apnea were protective factors.


    EHR‐based prediction model had good performance in identifying 5‐year MCI to ACD conversion and has potential to assist triaging of at‐risk patients.


    Of 59,782 veterans with mild cognitive impairment (MCI), 15,420 (25.8%) converted to all‐cause dementia within 5 years.

    Electronic health record prediction models demonstrated good performance (area under the receiver operating characteristic curve 0.73; Brier 0.18).

    Age and vascular‐related morbidities were predictors of dementia conversion.

    Synthetic data was comparable to real data in modeling MCI to dementia conversion.

    Key Points

    An electronic health record–based model using demographic and co‐morbidity data had good performance in identifying veterans who convert from mild cognitive impairment (MCI) to all‐cause dementia (ACD) within 5 years.

    Increased age, stroke, cerebrovascular disease, myocardial infarction, hypertension, and diabetes were risk factors for 5‐year conversion from MCI to ACD.

    High body mass index, alcohol abuse, and sleep apnea were protective factors for 5‐year conversion from MCI to ACD.

    Models using synthetic data, analogs of real patient data that retain the distribution, density, and covariance between variables of real patient data but are not attributable to any specific patient, performed just as well as models using real patient data. This could have significant implications in facilitating widely distributed computing of health‐care data with minimized patient privacy concern that could accelerate scientific discoveries.

    more » « less
  4. Accurate and explainable health event predictions are becoming crucial for healthcare providers to develop care plans for patients. The availability of electronic health records (EHR) has enabled machine learning advances in providing these predictions. However, many deep-learning-based methods are not satisfactory in solving several key challenges: 1) effectively utilizing disease domain knowledge; 2) collaboratively learning representations of patients and diseases; and 3) incorporating unstructured features. To address these issues, we propose a collaborative graph learning model to explore patient-disease interactions and medical domain knowledge. Our solution is able to capture structural features of both patients and diseases. The proposed model also utilizes unstructured text data by employing an attention manipulating strategy and then integrates attentive text features into a sequential learning process. We conduct extensive experiments on two important healthcare problems to show the competitive prediction performance of the proposed method compared with various state-of-the-art models. We also confirm the effectiveness of learned representations and model interpretability by a set of ablation and case studies.

    more » « less
  5. Abstract Objective

    The use of electronic health records (EHRs) for clinical risk prediction is on the rise. However, in many practical settings, the limited availability of task-specific EHR data can restrict the application of standard machine learning pipelines. In this study, we investigate the potential of leveraging language models (LMs) as a means to incorporate supplementary domain knowledge for improving the performance of various EHR-based risk prediction tasks.


    We propose two novel LM-based methods, namely “LLaMA2-EHR” and “Sent-e-Med.” Our focus is on utilizing the textual descriptions within structured EHRs to make risk predictions about future diagnoses. We conduct a comprehensive comparison with previous approaches across various data types and sizes.


    Experiments across 6 different methods and 3 separate risk prediction tasks reveal that employing LMs to represent structured EHRs, such as diagnostic histories, results in significant performance improvements when evaluated using standard metrics such as area under the receiver operating characteristic (ROC) curve and precision-recall (PR) curve. Additionally, they offer benefits such as few-shot learning, the ability to handle previously unseen medical concepts, and adaptability to various medical vocabularies. However, it is noteworthy that outcomes may exhibit sensitivity to a specific prompt.


    LMs encompass extensive embedded knowledge, making them valuable for the analysis of EHRs in the context of risk prediction. Nevertheless, it is important to exercise caution in their application, as ongoing safety concerns related to LMs persist and require continuous consideration.

    more » « less