Skip to main content
. Author manuscript; available in PMC: 2011 Aug 1.
Published in final edited form as: Neurobiol Aging. 2010 Jun 11;31(8):1429–1442. doi: 10.1016/j.neurobiolaging.2010.04.022

Table 1.

ADNI subjects and biomarkers included in each study. Here we outline the subject samples analyzed for different classification tests. Subjects are split into independent training and testing samples approximately in a 3:1 ratio, except for the smaller studies, to ensure correct evaluation of classifier performance. The 3:1 ratio is used in several machine learning studies such as Vemuri et al. (2008a). MRI denotes that a 1.5T MRI scan was available; BMI denotes body mass index. CSF denotes that CSF-derived biomarkers were available

Number of Subjects (Training + Testing)
Study Biomarkers AD MCI CN
1 MRI, Age, ApoE, Sex, BMI 158 (118 + 40) 264 (184 + 80) 213 (163 + 50)
2a MRI, Age, ApoE, CSF 77 (57 + 20) 158 (118 + 40) 93 (68 + 25)
2b MRI, Age, ApoE, PET-FDG 79 (59 + 20) 191 (146 + 45) 94 (74 + 20)
2c MRI, Age, ApoE, CSF, PET-FDG 40 (20 + 20) 83 (43 + 40) 43 (23 + 20)
3a MRI, Age, ApoE, CSF, PET-FDG - 64 (41 + 23) -
3b MRI, ApoE, PET-FDG - 129 (67 + 62) -