Cargando…
Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation
BACKGROUND: Scalable and accurate health outcome prediction using electronic health record (EHR) data has gained much attention in research recently. Previous machine learning models mostly ignore relations between different types of clinical data (ie, laboratory components, International Classifica...
Autores principales: | , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
JMIR Publications
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7136840/ https://www.ncbi.nlm.nih.gov/pubmed/32202503 http://dx.doi.org/10.2196/16374 |
_version_ | 1783518328027873280 |
---|---|
author | Rongali, Subendhu Rose, Adam J McManus, David D Bajracharya, Adarsha S Kapoor, Alok Granillo, Edgard Yu, Hong |
author_facet | Rongali, Subendhu Rose, Adam J McManus, David D Bajracharya, Adarsha S Kapoor, Alok Granillo, Edgard Yu, Hong |
author_sort | Rongali, Subendhu |
collection | PubMed |
description | BACKGROUND: Scalable and accurate health outcome prediction using electronic health record (EHR) data has gained much attention in research recently. Previous machine learning models mostly ignore relations between different types of clinical data (ie, laboratory components, International Classification of Diseases codes, and medications). OBJECTIVE: This study aimed to model such relations and build predictive models using the EHR data from intensive care units. We developed innovative neural network models and compared them with the widely used logistic regression model and other state-of-the-art neural network models to predict the patient’s mortality using their longitudinal EHR data. METHODS: We built a set of neural network models that we collectively called as long short-term memory (LSTM) outcome prediction using comprehensive feature relations or in short, CLOUT. Our CLOUT models use a correlational neural network model to identify a latent space representation between different types of discrete clinical features during a patient’s encounter and integrate the latent representation into an LSTM-based predictive model framework. In addition, we designed an ablation experiment to identify risk factors from our CLOUT models. Using physicians’ input as the gold standard, we compared the risk factors identified by both CLOUT and logistic regression models. RESULTS: Experiments on the Medical Information Mart for Intensive Care-III dataset (selected patient population: 7537) show that CLOUT (area under the receiver operating characteristic curve=0.89) has surpassed logistic regression (0.82) and other baseline NN models (<0.86). In addition, physicians’ agreement with the CLOUT-derived risk factor rankings was statistically significantly higher than the agreement with the logistic regression model. CONCLUSIONS: Our results support the applicability of CLOUT for real-world clinical use in identifying patients at high risk of mortality. |
format | Online Article Text |
id | pubmed-7136840 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | JMIR Publications |
record_format | MEDLINE/PubMed |
spelling | pubmed-71368402020-04-09 Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation Rongali, Subendhu Rose, Adam J McManus, David D Bajracharya, Adarsha S Kapoor, Alok Granillo, Edgard Yu, Hong J Med Internet Res Original Paper BACKGROUND: Scalable and accurate health outcome prediction using electronic health record (EHR) data has gained much attention in research recently. Previous machine learning models mostly ignore relations between different types of clinical data (ie, laboratory components, International Classification of Diseases codes, and medications). OBJECTIVE: This study aimed to model such relations and build predictive models using the EHR data from intensive care units. We developed innovative neural network models and compared them with the widely used logistic regression model and other state-of-the-art neural network models to predict the patient’s mortality using their longitudinal EHR data. METHODS: We built a set of neural network models that we collectively called as long short-term memory (LSTM) outcome prediction using comprehensive feature relations or in short, CLOUT. Our CLOUT models use a correlational neural network model to identify a latent space representation between different types of discrete clinical features during a patient’s encounter and integrate the latent representation into an LSTM-based predictive model framework. In addition, we designed an ablation experiment to identify risk factors from our CLOUT models. Using physicians’ input as the gold standard, we compared the risk factors identified by both CLOUT and logistic regression models. RESULTS: Experiments on the Medical Information Mart for Intensive Care-III dataset (selected patient population: 7537) show that CLOUT (area under the receiver operating characteristic curve=0.89) has surpassed logistic regression (0.82) and other baseline NN models (<0.86). In addition, physicians’ agreement with the CLOUT-derived risk factor rankings was statistically significantly higher than the agreement with the logistic regression model. CONCLUSIONS: Our results support the applicability of CLOUT for real-world clinical use in identifying patients at high risk of mortality. JMIR Publications 2020-03-23 /pmc/articles/PMC7136840/ /pubmed/32202503 http://dx.doi.org/10.2196/16374 Text en ©Subendhu Rongali, Adam J Rose, David D McManus, Adarsha S Bajracharya, Alok Kapoor, Edgard Granillo, Hong Yu. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 23.03.2020. https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. |
spellingShingle | Original Paper Rongali, Subendhu Rose, Adam J McManus, David D Bajracharya, Adarsha S Kapoor, Alok Granillo, Edgard Yu, Hong Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title | Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title_full | Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title_fullStr | Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title_full_unstemmed | Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title_short | Learning Latent Space Representations to Predict Patient Outcomes: Model Development and Validation |
title_sort | learning latent space representations to predict patient outcomes: model development and validation |
topic | Original Paper |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7136840/ https://www.ncbi.nlm.nih.gov/pubmed/32202503 http://dx.doi.org/10.2196/16374 |
work_keys_str_mv | AT rongalisubendhu learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT roseadamj learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT mcmanusdavidd learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT bajracharyaadarshas learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT kapooralok learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT granilloedgard learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation AT yuhong learninglatentspacerepresentationstopredictpatientoutcomesmodeldevelopmentandvalidation |