On Tue, 24 Mar 2009 12:42:19 -0700, Weiying Dai
<[log in to unmask]> wrote:
>Dear SPMers:
>
>You can choose "GMsca: grand mean scaling" when you build group
comparison in Basic Models (SPM button), then you have the option to choose
to "scaling of overall mean" to a value A.
>
>But I am just wondering how the overall mean value calculated from each
image. Does this calculated from all the 3D voxels (including the air regions) or
just the brain voxels? If that is calculated from the brain region, what is the
threshold to define the brain region.
From the last version of SPM2's spm_global.m:
% spm_global returns the mean counts integrated over all the
% slices from the volume
%
% The mean is estimated after discounting voxels outside the object
% using a criteria of greater than > (global mean)/8
So: it just uses "brain voxels," where "brain voxels" are defined to be those
with value > (mean over entire parallelpiped)/8.
So...it's a proportional threshold, with proportionality constant 0.125.
>Thanks in Advance,
>Weiying
>
>
>
|