Table 3. Performance of models using different feature selection methods and classifiers.

Method	Validation AUC	Sensitivity (%)	Specificity (%)	Testing AUC	Sensitivity (%)	Specificity (%)
mRMR + SVM	0.92 (0.80, 0.98)	86 (18/21)	85 (22/26)	0.85 (0.81, 0.89)	87 (91/105)	74 (169/229)
FF + SVM	0.90 (0.77, 0.97)	86 (18/21)	88 (23/26)	0.84 (0.80, 0.88)	83 (87/105)	75 (171/229)
Relief + RF	0.92 (0.80, 0.98)	81 (17/21)	92 (24/26)	0.86 (0.82, 0.90)	85 (89/105)	81 (186/229)
Relief + SVM	0.94 (0.85, 0.99)	86 (18/21)	92 (24/26)	0.88 (0.84, 0.91)	88 (92/105)	81 (186/229)