Dear Andy,
The tissue prior images you are using have regions where probabilities
are equal to zero, and this might create instabilities. Have a look to
this earlier post from John for things to try out:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=spm;a6060c5a.1609
Best regards,
Guillaume.
On 24/01/17 21:46, Andreas Horn wrote:
> Dear John & SPM experts,
>
> I tried applying the Unified Segmentation approach ("Segment" in SPM12, "New Segment" in 8) on standard anatomical images using the OASIS-30 priors (https://osf.io/rh9km/?action=download&version=1) but it failed giving me the error message below.
>
> Now the crucial difference between these priors is that they don't define tissue types but rather parts of the brain (cortex, subcortical structures, brainstem, cerebellum).
> I wonder if it's a bad idea in general to use Segment on these priors or whether there's a good and sensible idea to make this work.
>
> Thanks so much for any input!
>
> Andy
>
>
> Error message:
> ---------------------
>
> Failed 'Segment'
> Error using sqrtm (line 35)
> Expected input to be finite.
> In file "/Applications/MATLAB_R2015b.app/toolbox/matlab/matfun/sqrtm.m" (???), function "sqrtm" at line 35.
> In file "/Users/andreashorn/Documents/MATLAB/spm12/spm_preproc8.m" (v6798), function "spm_preproc8" at line 671.
> In file "/Users/andreashorn/Documents/MATLAB/spm12/spm_preproc_run.m" (v6365), function "run_job" at line 131.
> In file "/Users/andreashorn/Documents/MATLAB/spm12/spm_preproc_run.m" (v6365), function "spm_preproc_run" at line 41.
> In file "/Users/andreashorn/Documents/MATLAB/spm12/config/spm_cfg_preproc8.m" (v6798), function "spm_local_preproc_run" at line 418.
>
> The following modules did not run:
> Failed: Segment
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|