Dear Helmut,
thanks for looking into this - there was indeed a bug the way the clean
up option was dealt with, such that 'light' and 'thorough' were giving
the same output. The fix will be available in the next public update of
SPM12.
Best regards,
Guillaume.
On 24/02/16 11:27, MRI More wrote:
> Dear everyone,
>
> In the meantime I had a closer look at the clean-up issue, with SPM12 "Old Segment" there's a difference between light and thorough clean-up still, and a difference remains when using "Old Segment" together with the new tissue probability maps TPM.nii (I had wondered whether this would make a difference, as these TPMs have a "more diffuse" transition / voxels with very low but non-zero probabilities).
>
> So I looked at the code in spm_preproc_write ("Old") and spm_preproc_write8 ("New"), and I've noticed a difference with regard to the function clean_gwc . In "Old Segment", it's
>
> function [g,w,c] = clean_gwc(g,w,c,level)
> if nargin<4, level = 1; end
>
> In the newer version, it's
>
> function [P] = clean_gwc(P,level)
> if nargin<4, level = 1; end
>
> As it's just P and level as inputs it should read nargin<2 I guess. This way, one does indeed obtain a somewhat more extended clean-up with the "Thorough" setting, while the output with "Don't" and "Light" is identical to that obtained with nargin<4, see attachment. But maybe someone can verify this still, I'm far from understanding the code / why it is written the way it is ;-)
>
> Best
>
> Helmut
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|