TABLE 5.

10-fold cross-validation binary mean classification performance for MCI against AD using RELM classifier using different feature selection methods.

Feature selection method	Performance metrics	Mean (%)	Standard deviation	Max (%)	Min (%)
LASSO	Accuracy	93.86	0.766	94.90	89.128
	Sensitivity	91.93	0.757	93.67	89.836
	Specificity	95.92	1.204	96.61	88.580
	F-measure	0.968
FSASL	Accuracy	85.358	1.030	86.76	80.29
	Sensitivity	85.088	0.951	86.61	80.47
	Specificity	85.201	1.4227	86.829	79.27
	F-measure	0.879
LLCFS	Accuracy	90.32	1.06316	91.88	86.50
	Sensitivity	93.33	2.0782	95.63	87.43
	Specificity	87.49	0.9471	89.56	85.75
	F-measure	0.895
CFS	Accuracy	79.13	1.2768	81.595	75.10
	Sensitivity	83.59	1.2281	85.60	78.94
	Specificity	74.83	1.7084	78.927	70.54
	F-measure	0.795
SVMRFE	Accuracy	77.5974	0.8177	78.93	74.98
	Sensitivity	76.802	1.1299	79.225	73.26
	Specificity	78.182	0.8359	79.289	75.44
	F-measure	0.8169