Construction of risk prediction model of sarcopenia in senile patients with stroke based on Logistic regression and decision tree
KONG Linghui
YU Jie
ZHANG Huijun
CHEN Ping
Abstract：Objective:To explore the factors affecting sarcopenia in senile patients with stroke,construct risk prediction models,and evaluate their accuracy of prediction.Methods:A total of 489 senile patients with stroke from neurology department of a tertiary grade A hospital in Liaoning province were selected as the research subjects from September 2022 to April 2023.The risk prediction models of sarcopenia were constructed according to the results of Logistic regression analysis.The Nomogram and decision tree were painted,and the prediction efficiency of models were evaluated according to area under the curve(AUC)of receiver operator characteristic and confusion matrix.Results:The incidence of sarcopenia in senile patients with stroke was 37.6%.The results of logistic regression analysis show that smoking,age,activity of daily living(ADL),fall risk,nutrition and exercise habits were effect factors for sarcopenia in senile patients with stroke(P<0.05).The results of decision tree model showed that smoking,age,ADL,nutrition and exercise habits were decision-making factors for the sarcopenia in senile patients with stroke.The AUC of Logistic regression model was 0.959,and that of decision tree model training set and test set were 0.892 and 0.826,respectively.Conclusions:The Logistic regression model and decision tree model construct in this study have good predictive performance,which is helpful for clinical medical staff to screen the high-risk group of sarcopenia.
Keywords：senilestrokesarcopeniaLogistic regressiondecision treerehabilitationnursing
Publication Date：2024-05-25
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 1703-1710 )