Skip to main content

Table 4 Predictive performance of machine learning algorithms on ICU admission*

From: Implementation of a machine learning application in preoperative risk assessment for hip repair surgery

Algorithm

Accuracy

Sensitivity

Specificity

AUROC (95%CI)

Logistic Regression

0.791

0.792

0.791

0.856 (0.804–0.908)

Random Forest

0.760

0.750

0.760

0.844 (0.788–0.899)

SVM

0.706

0.708

0.706

0.730 (0.648–0.812)

KNN

0.658

0.542

0.662

0.630 (0.549–0.712)

lightGBM

0.769

0.771

0.769

0.842 (0.788–0.896)

MLP

0.734

0.812

0.731

0.829 (0.779–0.885)

XGBoost

0.709

0.708

0.709

0.825 (0.772–0.878)

  1. *ICU admission: 160 patients
  2. Abbreviations: ICU intensive care unit, AUROC area under receiver operating characteristic curve, CI confidence interval, SVM support vector machine, KNN K nearest neighbor, light GBM light gradient boosting machine, MLP multi-layer perception, XGBoost extreme gradient boosting, CI confidence interval