Prediction of Length of Hospital Stay of COVID-19 Patients Using Gradient Boosting Decision Tree.
Int J Biomater
; 2022: 6474883, 2022.
Article
in English
| MEDLINE | ID: covidwho-2038378
ABSTRACT
The aim of this paper is to predict the patient hospitalization time with coronavirus disease 2019 (COVID-19). It uses various data mining techniques, such as random forest. Many rules were derived by applying these techniques to the dataset. The extracted rules mainly were related to people over 55 years old. The rule with the most support states that if the person is between 70 and 80 years old, has cardiovascular disease, and the gender is female; then, the person will be hospitalized for at least five days. The gradient boosting random forest technique has performed better than other techniques. As a limitation of the study, it can be pointed out that a few features were unavailable and had not been recorded. Patients with diabetes, chronic respiratory problems, and cardiovascular diseases have a relatively long hospitalization. So, the hospital manager should consider a suitable priority for these patients. Older people were also more likely to take part in the selection rules.
Full text:
Available
Collection:
International databases
Database:
MEDLINE
Type of study:
Prognostic study
/
Randomized controlled trials
/
Reviews
Language:
English
Journal:
Int J Biomater
Year:
2022
Document Type:
Article
Affiliation country:
2022
Similar
MEDLINE
...
LILACS
LIS