Table 2.

Classification performance on the test set across the four partitions

Partition	AUC	Sensitivity	Specificity	PPV	NPV	F1-score
1	0.8155 (0.6813, 0.9429)	0.7143 (0.4667, 0.9333)	0.9167 (0.8472, 0.9722)	0.6250 (0.3750, 0.8571)	0.9429 (0.8873, 0.9863)	0.7981 (0.6729, 0.9016)
2	0.9067 (0.8600, 0.9494)	1.0000 (1.0000, 1.0000)	0.8133 (0.7260, 0.8987)	0.3636 (0.1739, 0.5714)	1.0000 (1.0000, 1.0000)	0.7152 (0.5798, 0.8362)
3	0.7899 (0.6416, 0.9293)	0.6667 (0.3846, 0.9286)	0.9130 (0.8438, 0.9726)	0.5714 (0.3000, 0.8235)	0.9403 (0.8788, 0.9857)	0.7709 (0.6318, 0.8889)
4	0.9016 (0.8500, 0.9508)	1.0000 (1.0000, 1.0000)	0.8033 (0.6964, 0.9000)	0.3684 (0.1500, 0.6000)	1.0000 (1.0000, 1.0000)	0.7147 (0.5662, 0.8383)
x˜	0.8586	0.8571	0.8632	0.4699	0.9714	0.7431

Data in parentheses are 95% confidence intervals

PPV positive predictive value, NPV negative predictive value, AUC area under the receiver operating characteristic curve