Table 2.

Evaluation of the performance of the four models.

		AUC(95%CI)	Accuracy(95%CI)	Sensitivity(95%CI)	Specificity(95%CI)	F1 Score(95%CI)
KNN	Training set	0.973 (0.954–0.992)	0.954 (0.950-0.958)	0.946 (0.922-0.969)	0.910 (0.889-0.931)	0.883 (0.865-0.900)
	Validation set	0.945 (0.888–0.999)	0.949 (0.938–0.959)	0.879 (0.833–0.925)	0.931 (0.900–0.963)	0.838 (0.795–0.881)
XGBoost	Training set	0.987 (0.978–0.995)	0.934 (0.928–0.939)	0.957 (0.948–0.965)	0.927 (0.920–0.933)	0.799 (0.786–0.813)
	Validation set	0.963 (0.922–1.000)	0.916 (0.902–0.929)	0.926 (0.888–0.964)	0.923 (0.903–0.943)	0.743 (0.699–0.787)
RandomForest	Training set	0.968 (0.951–0.985)	0.897 (0.885–0.909)	0.925 (0.902–0.948)	0.890 (0.872–0.908)	0.714 (0.692–0.735)
	Validation set	0.961 (0.918–0.999)	0.889 (0.868–0.910)	0.890 (0.852–0.927)	0.961 (0.945–0.976)	0.678 (0.622–0.735)
SVM	Training set	0.967 (0.946–0.989)	0.911 (0.905–0.916)	0.923 (0.910–0.935)	0.906 (0.898–0.914)	0.740 (0.731–0.749)
	Validation set	0.962 (0.917–1.000)	0.897 (0.880–0.914)	0.909 (0.869–0.950)	0.951 (0.932–0.970)	0.697(0.644–0.750)

CI, confidence interval.