Table 2.

Performance of the six models at the best threshold in the k-fold set.

Variables	AUC (95% CI)	Specificity (95% CI)	Sensitivity (95% CI)	Accuracy (95% CI)
RF	0.62 (0.40–0.83)	0.48 (0.19–1.00)	1.00 (0.38–1.00)	0.57 (0.37–0.86)
KNN	0.61 (0.38–0.84)	0.70 (0.15–0.89)	0.75 (0.38–1.00)	0.71 (0.34–0.86)
LOG	0.71 (0.52–0.89)	0.65 (0.30–0.91)	0.82 (0.45–1.00)	0.71 (0.53–0.85)
SVM	0.74 (0.55–0.93)	0.74 (0.52–0.96)	0.82 (0.55–1.00)	0.76 (0.62–0.91)
XGB	0.57 (0.40–0.74)	0.57 (0.00–1.00)	0.57 (0.00–1.00)	0.60 (0.40–0.74)
GBM	0.67 (0.50- 0.84)	0.63 (0.19–0.89)	0.79 (0.43–1.00)	0.68 (0.46–0.80)

AUC, area under the curve; CI, confidence interval; GBM, light gradient Boosting Machine; KNN, K-nearest neighbor; LOG, logistic regression; RAF, random forest; SVM, support vector machine; XGB, extreme Gradient Boosting.