Table 2.

Predictive perfrmance of ANN, SVM and NBC models.

Statistics*	ANN	SVM	P-value (ANN vs. SVM)	Naïve Bayes	P-value (ANN vs. NBC)	P-value (overall comparison)
Sensitivity	75% (63.3–83.3%)	62.5% (50–62.5%)	<0.001	62.5% (50–75%)	0.001	<0.001
Specificity	75% (62.5–83.3%)	75% (50–87.5%)	0.081	75% (62.5–75%)	0.121	0.151
Accuracy	75% (68.8–76.6%)	62.5% (56.3–68.8%)	<0.001	62.5% (56.3–68.8%)	<0.001	<0.001
C statistic	0.77 (0.68–0.84)	0.63 (0.56–0.69)	<0.001	0.63 (0.56–0.69)	<0.001	<0.001

^*presented with medians (IQR).

ANN indicates artificial neural network; SVM indicates support vector machine; NBC indaictaes Naïve Bayes classifier.