Published online Apr 26, 2023. doi: 10.12998/wjcc.v11.i12.2716
Peer-review started: November 1, 2022
First decision: January 30, 2023
Revised: February 12, 2023
Accepted: March 17, 2023
Article in press: March 17, 2023
Published online: April 26, 2023
coronavirus disease 2019 (COVID-19) is a global pandemic that requires early identification and intervention to reduce morbidity and mortality. Chest computed tomography (CT) score has been shown to be a factor in the diagnosis and treatment of COVID-19 pneumonia. However, there is currently a lack of effective early warning systems for severe/critical COVID-19.
To develop a severe/critical COVID-19 prediction model using a combination of imaging scores, clinical features, and biomarker levels.
To identify key factors in predicting severe/critical COVID-19 cases using improved chest CT scores and machine learning algorithms.
The study used an improved scoring system to extract chest CT characteristics of COVID-19 patients, and considered general clinical indicators such as dyspnea, oxygen saturation, alanine aminotransferase, and aspartate aminotransferase. Lasso regression was employed to evaluate the significance of different disease characteristics.
A COVID-19 severe/critical early warning system was established using machine learning algorithms including XGBClassifier, Logistic Regression, MLPClassifier, RandomForestClassifier, and AdaBoost Classifier.
The prediction model based on improved CT scores and machine learning algorithms is effective in detecting early warning signals of severe/critical COVID-19.
The findings suggest that this method is a feasible solution for early detection of severe/critical COVID-19 evolution and may help reduce morbidity and mortality.