Cargando…
61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19
BACKGROUND: As the ongoing COVID-19 pandemic develops, there is a need for prediction rules to guide clinical decisions. Previous reports have identified risk factors using statistical inference model. The primary goal of these models is to characterize the relationship between variables and outcome...
Autores principales: | , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Oxford University Press
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7778070/ http://dx.doi.org/10.1093/ofid/ofaa439.371 |
_version_ | 1783631051559534592 |
---|---|
author | Rodriguez-Nava, Guillermo Trelles-Garcia, Daniela Patricia Yanez-Bello, Maria Adriana Chung, Chul Won Chaudry, Sana Khan, Aimen Friedman, Harvey Hines, David |
author_facet | Rodriguez-Nava, Guillermo Trelles-Garcia, Daniela Patricia Yanez-Bello, Maria Adriana Chung, Chul Won Chaudry, Sana Khan, Aimen Friedman, Harvey Hines, David |
author_sort | Rodriguez-Nava, Guillermo |
collection | PubMed |
description | BACKGROUND: As the ongoing COVID-19 pandemic develops, there is a need for prediction rules to guide clinical decisions. Previous reports have identified risk factors using statistical inference model. The primary goal of these models is to characterize the relationship between variables and outcomes, not to make predictions. In contrast, the primary purpose of machine learning is obtaining a model that can make repeatable predictions. The objective of this study is to develop decision rules tailored to our patient population to predict ICU admissions and death in patients with COVID-19. METHODS: We used a de-identified dataset of hospitalized adults with COVID-19 admitted to our community hospital between March 2020 and June 2020. We used a Random Forest algorithm to build the prediction models for ICU admissions and death. Random Forest is one of the most powerful machine learning algorithms; it leverages the power of multiple decision trees, randomly created, for making decisions. RESULTS: 313 patients were included; 237 patients were used to train each model, 26 were used for testing, and 50 for validation. A total of 16 variables, selected according to their availability in the Emergency Department, were fit into the models. For the survival model, the combination of age >57 years, the presence of altered mental status, procalcitonin ≥3.0 ng/mL, a respiratory rate >22, and a blood urea nitrogen >32 mg/dL resulted in a decision rule with an accuracy of 98.7% in the training model, 73.1% in the testing model, and 70% in the validation model (Table 1, Figure 1). For the ICU admission model, the combination of age < 82 years, a systolic blood pressure of ≤94 mm Hg, oxygen saturation of ≤93%, a lactate dehydrogenase >591 IU/L, and a lactic acid >1.5 mmol/L resulted in a decision rule with an accuracy of 99.6% in the training model, 80.8% in the testing model, and 82% in the validation model (Table 2, Figure 2). Table 1. Measures of Performance in Predicting Inpatient Mortality [Image: see text] CONCLUSION: We created decision rules using machine learning to predict ICU admission or death in patients with COVID-19. Although there are variables previously described with statistical inference, these decision rules are customized to our patient population; furthermore, we can continue to train the models fitting more data with new patients to create even more accurate prediction rules. Figure 1. Receiver Operating Characteristic (ROC) Curve for Inpatient Mortality [Image: see text] Table 2. Measures of Performance in Predicting Intensive Care Unit Admission [Image: see text] Figure 2. Receiver Operating Characteristic (ROC) Curve for Intensive Care Unit Admission [Image: see text] DISCLOSURES: All Authors: No reported disclosures |
format | Online Article Text |
id | pubmed-7778070 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | Oxford University Press |
record_format | MEDLINE/PubMed |
spelling | pubmed-77780702021-01-07 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 Rodriguez-Nava, Guillermo Trelles-Garcia, Daniela Patricia Yanez-Bello, Maria Adriana Chung, Chul Won Chaudry, Sana Khan, Aimen Friedman, Harvey Hines, David Open Forum Infect Dis Poster Abstracts BACKGROUND: As the ongoing COVID-19 pandemic develops, there is a need for prediction rules to guide clinical decisions. Previous reports have identified risk factors using statistical inference model. The primary goal of these models is to characterize the relationship between variables and outcomes, not to make predictions. In contrast, the primary purpose of machine learning is obtaining a model that can make repeatable predictions. The objective of this study is to develop decision rules tailored to our patient population to predict ICU admissions and death in patients with COVID-19. METHODS: We used a de-identified dataset of hospitalized adults with COVID-19 admitted to our community hospital between March 2020 and June 2020. We used a Random Forest algorithm to build the prediction models for ICU admissions and death. Random Forest is one of the most powerful machine learning algorithms; it leverages the power of multiple decision trees, randomly created, for making decisions. RESULTS: 313 patients were included; 237 patients were used to train each model, 26 were used for testing, and 50 for validation. A total of 16 variables, selected according to their availability in the Emergency Department, were fit into the models. For the survival model, the combination of age >57 years, the presence of altered mental status, procalcitonin ≥3.0 ng/mL, a respiratory rate >22, and a blood urea nitrogen >32 mg/dL resulted in a decision rule with an accuracy of 98.7% in the training model, 73.1% in the testing model, and 70% in the validation model (Table 1, Figure 1). For the ICU admission model, the combination of age < 82 years, a systolic blood pressure of ≤94 mm Hg, oxygen saturation of ≤93%, a lactate dehydrogenase >591 IU/L, and a lactic acid >1.5 mmol/L resulted in a decision rule with an accuracy of 99.6% in the training model, 80.8% in the testing model, and 82% in the validation model (Table 2, Figure 2). Table 1. Measures of Performance in Predicting Inpatient Mortality [Image: see text] CONCLUSION: We created decision rules using machine learning to predict ICU admission or death in patients with COVID-19. Although there are variables previously described with statistical inference, these decision rules are customized to our patient population; furthermore, we can continue to train the models fitting more data with new patients to create even more accurate prediction rules. Figure 1. Receiver Operating Characteristic (ROC) Curve for Inpatient Mortality [Image: see text] Table 2. Measures of Performance in Predicting Intensive Care Unit Admission [Image: see text] Figure 2. Receiver Operating Characteristic (ROC) Curve for Intensive Care Unit Admission [Image: see text] DISCLOSURES: All Authors: No reported disclosures Oxford University Press 2020-12-31 /pmc/articles/PMC7778070/ http://dx.doi.org/10.1093/ofid/ofaa439.371 Text en © The Author 2020. Published by Oxford University Press on behalf of Infectious Diseases Society of America. http://creativecommons.org/licenses/by-nc-nd/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com |
spellingShingle | Poster Abstracts Rodriguez-Nava, Guillermo Trelles-Garcia, Daniela Patricia Yanez-Bello, Maria Adriana Chung, Chul Won Chaudry, Sana Khan, Aimen Friedman, Harvey Hines, David 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title | 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title_full | 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title_fullStr | 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title_full_unstemmed | 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title_short | 61. Using Machine Learning for Prediction of Poor Clinical Outcomes in Adult Patients Hospitalized with COVID-19 |
title_sort | 61. using machine learning for prediction of poor clinical outcomes in adult patients hospitalized with covid-19 |
topic | Poster Abstracts |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7778070/ http://dx.doi.org/10.1093/ofid/ofaa439.371 |
work_keys_str_mv | AT rodrigueznavaguillermo 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT trellesgarciadanielapatricia 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT yanezbellomariaadriana 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT chungchulwon 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT chaudrysana 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT khanaimen 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT friedmanharvey 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 AT hinesdavid 61usingmachinelearningforpredictionofpoorclinicaloutcomesinadultpatientshospitalizedwithcovid19 |