Table 3.

Algorithm comparisons of the proposed method with the related works for ADAS-Cog and MMSE prediction.

Method	Target	Subject	Feature	Result (Corr)
Duchesne et al., 2009	M12 MMSE	75 NC, 49 MCI, 75 AD	Baseline MRI, age, gender, years of education	MMSE: 0.31 (p = 0.03)
Stonnington et al., 2010	Baseline ADAS-Cog and MMSE	Set 1:73 AD, 91 NC Set 2: (ADNI) 113 AD, 351 MCI, 122 NC	BaselineMRI, CSF	MMSE Set1: 0.7 (p < 10e-5) Set 2: 0.48 (p < 10e-5) ADAS-Cog Set 2: 0.57 (p < 10e-5)
Zhang et al., 2012	Baseline,M06,M12,M24 ADAS-Cog, MMSE	ADNI: 91 AD, 202 MCI, 152 NC	Baseline,M06, M12, M24 MRI	Average MMSE: 0.613 (p < 10e-5) Average ADAS-Cog: 0.639 (p < 10e-5)
Proposed	Baseline, M06, M12, M24 ADAS-Cog, MMSE	ADNI: 91 AD, 202 MCI, 152 NC	Baseline MRI	Average MMSE: 0.7538 (p < 10e-5) Average ADAS-Cog: 0.7875 (p < 10e-5)

AD, MCI, and NC denote Alzheimer's disease patients, mild cognitive impairment patients, and normal controls, respectively.