TABLE 3.

The classification performance comparison of the five methods.

Group	Methods	ACC (%) ± SD	SEN (%) ± SD	SPE (%) ± SD	GMean (%) ± SD	F1 (%) ± SD
AD-HC	Lasso + SVM	85.45 ± 1.10	75.34 ± 2.87	91.89 ± 0.81	83.19 ± 1.49	80.05 ± 1.66
	GroupLasso + SVM	86.16 ± 0.78	77.13 ± 1.73	91.97 ± 1.00	84.24 ± 0.93	80.75 ± 2.75
	ElasticNet + SVM	84.56 ± 1.00	74.90 ± 2.14	90.72 ± 1.24	82.41 ± 1.13	79.04 ± 1.24
	Sparse Group Lasso + SVM	85.75 ± 0.83	75.99 ± 2.29	92.20 ± 0.88	83.69 ± 1.11	80.89 ± 1.18
	Multi-modal LassoNet	90.68 ± 0.34	88.81 ± 0.68	91.91 ± 0.55	90.34 ± 0.36	88.25 ± 0.52
AD-EMCI	Lasso + SVM	75.88 ± 0.58	93.06 ± 0.87	54.22 ± 0.95	71.03 ± 0.61	81.15 ± 0.56
	GroupLasso + SVM	75.92 ± 1.04	93.12 ± 0.63	54.33 ± 1.77	71.12 ± 1.14	81.16 ± 0.87
	ElasticNet + SVM	76.13 ± 0.61	92.91 ± 0.94	54.98 ± 1.96	71.45 ± 1.01	81.27 ± 0.60
	Sparse Group Lasso + SVM	70.23 ± 0.63	90.44 ± 1.05	44.69 ± 1.82	63.56 ± 1.04	77.05 ± 0.67
	Multi-modal LassoNet	83.63 ± 0.74	87.32 ± 1.22	79.00 ± 1.33	83.05 ± 0.76	85.70 ± 0.84
EMCI-HC	Lasso + SVM	67.04 ± 0.69	78.67 ± 1.98	57.61 ± 1.49	67.30 ± 0.68	68.12 ± 0.90
	GroupLasso + SVM	84.42 ± 0.65	96.62 ± 0.65	74.43 ± 0.98	84.80 ± 0.52	84.08 ± 0.57
	ElasticNet + SVM	83.76 ± 0.42	94.69 ± 0.69	74.72 ± 0.92	84.11 ± 0.44	84.07 ± 0.41
	Sparse Group Lasso + SVM	83.20 ± 1.15	95.83 ± 0.88	72.74 ± 1.43	83.49 ± 1.14	70.86 ± 1.16
	Multi-modal LassoNet	88.77 ± 0.70	90.87 ± 1.05	87.06 ± 0.95	88.94 ± 0.69	87.92 ± 0.83