Table 3.

Evaluation of the performance of the four models in the internal validation set

		AUC(95%CI)	Accuracy(95%CI)	Sensitivity(95%CI)	Specificity(95%CI)	F1 Score(95%CI)
KNN	Training set	0.948 (0.914–0.983)	0.944 (0.941–0.947)	0.963 (0.952–0.975)	0.889 (0.878–0.899)	0.779 (0.761–0.798)
	Validation set	0.796 (0.637–0.949)	0.932 (0.922–0.942)	0.674 (0.565–0.783)	0.898 (0.868–0.928)	0.646 (0.564–0.727)
XGBoost	Training set	0.976 (0.960–0.992)	0.943 (0.934–0.951)	0.913 (0.904–0.921)	0.946 (0.937–0.954)	0.716 (0.690–0.742)
	Validation set	0.906 (0.830–0.979)	0.903 (0.886–0.921)	0.922 (0.882–0.961)	0.828 (0.775–0.880)	0.569 (0.459–0.678)
RF	Training set	0.880 (0.818–0.941)	0.818 (0.783–0.854)	0.826 (0.785–0.868)	0.815 (0.775–0.854)	0.417 (0.377–0.457)
	Validation set	0.842 (0.723–0.959)	0.808 (0.772–0.844)	0.804 (0.731–0.877)	0.822 (0.775–0.869)	0.393 (0.352–0.433)
SVM	Training set	0.959 (0.933–0.985)	0.872 (0.816–0.928)	0.916 (0.858–0.973)	0.864 (0.797–0.931)	0.591 (0.468–0.715)
	Validation set	0.851 (0.730–0.970)	0.833 (0.792–0.874)	0.849 (0.754–0.945)	0.813 (0.764–0.863)	0.445 (0.356–0.535)

CI, confidence interval; KNN, k-nearest neighbor; XGBoost, extreme gradient boosting; RF, random forest; SVM, support vector machine; AUC, area under the curve