Skip to main content
. Author manuscript; available in PMC: 2013 Jan 24.
Published in final edited form as: Neuroimage. 2011 Jan 31;56(1):185–196. doi: 10.1016/j.neuroimage.2011.01.062

Table 3.

The mean classification rate (the highest value of the column in bold) and standard deviation for the ADNI N = 321 cohort using both semi-automatic and automatic segmentations with and without partial volume (PV) correction. The results are reported both for the training and test sets. The difference between semi-automatically and automatically (both with and without PV correction) generated volumes is statistically significant for all columns (not indicated in the table).

Classification rate CN(N = 89)–AD(N = 82) SMCI(N = 76)–PMCI
(N = 75)
ADNI (N = 321) Training set Test set Training set Test set
Semi-automatic volumes 85.5 ± 1.9 82.5 ± 4.4 72.9 ± 2.5 71.4 ± 5.4
Automatic volumes — No PV 84.5 ± 2.0 82.1 ± 4.5 68.5 ± 2.5 63.7 ± 5.8
Automatic volumes — PV 84.7 ± 2.0* 83.4 ± 4.4* 68.9 ± 2.5* 64.9 ± 6.1*

The statistically significant difference with and without partial volume correction is shown by ‘*’.