Table 3.

Classification performances of the different cross-validation methods.

CV methods	Dataset	AUC (95% CI)	Accuracy (%)	Sensitivity (%; 95% CI)	Specificity (%; 95% CI)
10-fold	Primary	0.974 (0.946–1)	93.8	93.8 (82.8–98.69)	90.6 (74.98–98.02)
	Validation	0.980 (0.929–1)	90.0	90.9 (58.72–99.77)	88.9 (51.75–99.72)
5-fold	Primary	0.978 (0.953–1)	90.0	97.9 (88.93–99.95)	78.1 (60.03–90.72)
	Validation	0.981 (0.925–1)	90.0		77.8 (39.99–97.19)
LOO	Primary	0.971 (0.940–1)	83.8	75.0 (60.4–86.36)	96.9 (83.78–99.92)
	Validation	0.968 (0.918–1)	85.0	72.7 (39.03–93.98)

CV, cross-validation; LOO, leave one out; AUC, area under the curve; CI, confidence interval.