Table 2.

Performance of four different classifiers with three different feature selection methods in the modeling set

Classifier	Feature selection	AUC	Accuracy	Specificity	Sensitivity
LDA	WRST	0.74 ± 0.07	0.83 ± 0.03	0.91 ± 0.10	0.67 ± 0.04
	MRMR	0.83 ± 0.05	0.79 ± 0.07	0.87 ± 0.08	0.65 ± 0.09
	RF	0.77 ± 0.05	0.81 ± 0.03	0.91 ± 0.10	0.62 ± 0.06
QDA	WRST	0.87 ± 0.02	0.88 ± 0.02	0.93 ± 0.10	0.78 ± 0.04
	MRMR	0.81 ± 0.04	0.84 ± 0.05	0.88 ± 0.14	0.76 ± 0.04
	RF	0.83 ± 0.06	0.85 ± 0.3	0.91 ± 0.15	0.72 ± 0.06
RF	WRST	0.81 ± 0.05	0.77 ±0.04	0.87 ± 0.06	0.59 ± 0.02
	MRMR	0.84 ± 0.03	0.81 ±0.04	0.87 ± 0.06	0.68 ± 0.06
	RF	0.78 ± 0.04	0.74 ± 0.04	0.83 ± 0.13	0.58 ± 0.05
SVM	WRST	0.86 ± 0.02	0.82 ± 0.03	0.93 ± 0.06	0.62 ± 0.04
	MRMR	0.79 ± 0.07	0.72 ± 0.04	0.90 ± 0.15	0.35 ± 0.06
	RF	0.84 ± 0.02	0.79 ± 0.02	0.92 ± 0.08	0.53 ± 0.03

Abbreviations: AUC, area under curve; LDA/QDA, linear/quadratic discriminant analysis; MRMR, minimum redundancy, maximum relevance feature selection method; RF, random forest; SVM, support vector machine; WRST, Wilcoxon rank sum test.

The best performance in each metric/column is shown in bold.