Prediction of COVID-19 Mortality in the Intensive Care Unit Using Machine Learning.
Stud Health Technol Inform
; 302: 536-540, 2023 May 18.
Artículo
en Inglés
| MEDLINE | ID: covidwho-2326002
ABSTRACT
Since its emergence, the COVID-19 pandemic still poses a major global health threat. In this setting, a number of useful machine learning applications have been explored to assist clinical decision-making, predict the severity of disease and admission to the intensive care unit, and also to estimate future demand for hospital beds, equipment, and staff. The present study examined demographic data, hematological and biochemical markers routinely measured in Covid-19 patients admitted to the intensive care unit (ICU) of a public tertiary hospital, in relation to the ICU outcome, during the second and third Covid-19 waves, from October 2020 until February 2022. In this dataset, we applied eight well-known classifiers of the caret package for machine learning of the R programming language, to evaluate their performance in forecasting ICU mortality. The best performance regarding area under the receiver operating characteristic curve (AUC-ROC) was observed with Random Forest (0.82), while k-nearest neighbors (k-NN) were the lowest performing machine learning algorithm (AUC-ROC 0.59). However, in terms of sensitivity, XGB outperformed the other classifiers (max Sens 0.7). The six most important predictors of mortality in the Random Forest model were serum urea, age, hemoglobin, C-reactive protein, platelets, and lymphocyte count.
Palabras clave
Texto completo:
Disponible
Colección:
Bases de datos internacionales
Base de datos:
MEDLINE
Asunto principal:
COVID-19
Tipo de estudio:
Estudio experimental
/
Estudio observacional
/
Estudio pronóstico
/
Ensayo controlado aleatorizado
Límite:
Humanos
Idioma:
Inglés
Revista:
Stud Health Technol Inform
Asunto de la revista:
Informática Médica
/
Investigación sobre Servicios de Salud
Año:
2023
Tipo del documento:
Artículo
País de afiliación:
Shti230200
Similares
MEDLINE
...
LILACS
LIS