Table 2.

Classification performance of the proposed model a) for Score 2 vs. Score 3 vs. Score 4 vs. Score 5 classification using 5-fold cross-validation.

Classifiers	Sensitivity (%)	Specificity (%)	Accuracy (%)	AUC
LDA	85.50 ± 1.95	75.00 ± 1.10	88.00 ± 0.98	0.94
Linear SVM	81.75 ± 2.70	74.30 ± 2.25	83.10 ± 1.10	0.91
Gaussian SVM	86.00 ± 1.50	74.50 ± 1.70	86.40 ± 1.65	0.92
b) for Low score (2 and 3) vs. High score (4 and 5) classification using fivefold cross validation
Classifiers	Sensitivity (%)	Specificity (%)	Accuracy (%)	AUC
LDA	90.00 ± 2.80	81.25 ± 1.95	89.80 ± 0.95	0.95
Linear SVM	91.45 ± 3.65	95.85 ± 1.25	93.20 ± 2.10	0.99
Gaussian SVM	88.60 ± 2.05	91.67 ± 2.50	86.40 ± 1.90	0.96