Prediction Length of Stay in Intensive Care Unit in the Presence of Missing Data

Authors : Zeliha ERGUL AYDIN, Zehra KAMIŞLI ÖZTÜRK

Pages : 48-53

View : 16 | Download : 12

Publication Date : 2021-09-30

Article Type : Research Paper

Abstract :Patients’ length of stay insert ignore into journalissuearticles values(LOS); in intensive care units insert ignore into journalissuearticles values(ICU); is an important factor for managing limited ICU resources such as beds, staffing, medicines, and medical devices. The goal of this study predicts that the ICU length of stay of patients is more than 3 days or not with Support Vector Machine insert ignore into journalissuearticles values(SVM);, Logistic Regression insert ignore into journalissuearticles values(LR);, XGBoost classifiers. We retrieved the 37,600 ICU patients’ demographics data and last measured vital signs in their first 12 hours of stay from the MIMIC-III database. We filled the missing patients` data with three missing data imputation methods, namely k-nearest neighbor imputation insert ignore into journalissuearticles values(KNN);, multivariate imputation by chained equations insert ignore into journalissuearticles values(MICE);, and SoftImpute. Our results indicated that filling missing data with the Soft-Impute yielded the highest AUC score for all classifiers. We obtained the highest area under the curve score as 66.1% with the XGBoost classifier and Soft-Impute missing data imputation.
Keywords : length of stay prediction, classification algorithms, intensive care unit, MIMIC III