Research on risk prediction model for unplanned return to ICU based on machine learning algorithm.
Objective: To construct a risk prediction model for unplanned return to intensive care unit (ICU) based on machine learning algorithm. Methods: A total of 3 250 ICU patients from a tertiary grade A hospital in Shanxi province from October 12, 2019 to May 21, 2023 were selected as the research subjec...
| Publicado en: | Chinese Nursing Research Vol. 38; no. 22; pp. 3976 - 3983 |
|---|---|
| Autores principales: | , , , , |
| Formato: | research tables/charts Journal Article |
| Publicado: |
Chinese Nursing Research Editorial Office
Nov2024
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=181634762&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 181634762 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 10096493 YV6 jtl: Chinese Nursing Research issn: 10096493 maglogo: N pubinfo: dt: Nov2024 vid: 38 iid: 22 pid: 37375 pub: Chinese Nursing Research Editorial Office artinfo: ui: 181634762 181634762 181634762 10.12102/j.issn.1009-6493.2024.22.004 181634762 ppf: 3976 ppct: 7 formats: tig: atl: Research on risk prediction model for unplanned return to ICU based on machine learning algorithm. aug: au: LI Mengke SUN Yan LIU Hongqi QU Jingchen HOU Ruiqin affil: Shanxi Medical University, Shanxi 030606 China sug: subj: Risk Assessment Prediction Models Machine Learning Algorithms Intensive Care Units Patient Admission Human Male Female China Models, Statistical Tertiary Health Care China Descriptive Statistics ROC Curve Logistic Regression Decision Trees Chronic Disease Prognosis Hospitalization Body Temperature Length of Stay Creatinine Blood Heart Rate Neutrophils Diastolic Pressure Systolic Pressure Glasgow Coma Scale Scales Male Female ab: Objective: To construct a risk prediction model for unplanned return to intensive care unit (ICU) based on machine learning algorithm. Methods: A total of 3 250 ICU patients from a tertiary grade A hospital in Shanxi province from October 12, 2019 to May 21, 2023 were selected as the research subjects. A risk prediction model for return to ICU was constructed based on multiple machine learning algorithms, and the performance of the models was compared. Analyze the importance ranking of each variable based on the best performing model. Results: The light gradient boosting machine had the best comprehensive performance, with area under the receiver operating characteristic curve (AUROC) of 0.996 8, followed by random forest (AUROC = 0.996 4), gradient boosting decision tree (AUROC = 0. 992 4), adaptive boosting (AUROC = 0.953 0), and Logistic regression (AUROC = 0.814 5). The top 15 variables in the importance ranking based on light gradient boosting machine were K+, blood loss, score of Glasgow Coma Scale, score of Acute Physiology and Chronic Health Evaluation II, Na+, CRP, alcohol consumption history, minimum body temperature, ICU stay time, blood creatinine, minimum heart rate, neutrophil count, minimum diastolic blood pressure, bicarbonate, and maximum systolic blood pressure. Conclusion: The risk prediction models for unplanned return to ICU based on machine learning algorithm performs well. Researchers can use this type of algorithm to establish the risk prediction model to identify high-risk patients, provide targeted intervention measures, and improve the quality of healthcare. pubtype: Academic Journal doctype: research tables/charts Journal Article ougenre: Article language: Chinese refInfo: holdings: @attributes: islocal: N |
|---|