Table 2.

Comparison of ADAS-Cog and MMSE prediction results by our proposed SMTL method and three feature selection methods: Lasso, TGL, and cFSGL, using Corr and RMSE measurement.

	ADAS–Cog				MMSE
	Lasso	TGL	cFSGL	SMTL	Lasso	TGL	cFSGL	SMTL
Baseline Corr	0.32 ± 0.08	0.49 ± 0.08	0.58 ± 0.08	0.77 ± 0.05	0.51 ± 0.09	0.56 ± 0.12	0.64 ± 0.08	0.75 ± 0.08
M06 Corr	0.48 ± 0.07	0.50 ± 0.07	0.59 ± 0.05	0.78 ± 0.04	0.49 ± 0.07	0.48 ± 0.10	0.63 ± 0.10	0.79 ± 0.10
M12 Corr	0.46 ± 0.12	0.52 ± 0.09	0.62 ± 0.05	0.79 ± 0.04	0.52 ± 0.06	0.57 ± 0.08	0.64 ± 0.12	0.79 ± 0.12
M24 Corr	0.57 ± 0.08	0.53 ± 0.08	0.69 ± 0.05	0.84 ± 0.04	0.61 ± 0.05	0.58 ± 0.05	0.65 ± 0.05	0.83 ± 0.06
Baseline RMSE	5.19 ± 0.33	5.53 ± 0.32	5.20 ± 0.33	3.81 ± 0.45	2.25 ± 0.20	2.30 ± 0.24	2.24 ± 0.21	1.75 ± 0.20
M06 RMSE	5.48 ± 0.62	5.90 ± 0.65	5.48 ± 0.63	4.36 ± 0.46	2.75 ± 0.31	2.85 ± 0.33	2.76 ± 0.32	2.31 ± 0.29
M12 RMSE	6.25 ± 0.58	6.74 ± 0.63	6.24 ± 0.59	4.91 ± 0.56	3.42 ± 0.49	3.59 ± 0.52	3.42 ± 0.50	2.48 ± 0.40
M24 RMSE	7.95 ± 1.20	7.81 ± 1.39	7.96 ± 1.23	6.00 ± 1.00	4.05 ± 0.58	4.48 ± 0.82	4.04 ± 0.58	3.00 ± 0.38