[MINC-users] nonlinear registration bombs with 0.5 mm isotropic voxels

[Andrew Janke]

Hi Patrick

On Thu, May 28, 2009 at 06:33, Patrick Bermudez
<patrick at bic.mni.mcgill.ca> wrote:
> I am experiencing a problem with a nonlinear fit of 0.5 mm isotropic
> volumes.
>
> ...
>
> Iteration  1 of 20
> Died at ./04mm_NLfit.pl line 285.

My first punt is that with 0.5mm voxels you are running out of memory.
With that in mind I would run minctracc with the -no_super flag. The
general rule of thumb is that you should use -no_super for anything
less than 1mm on "head" data.

--
Andrew Janke
(a.janke at gmail.com || http://a.janke.googlepages.com/)
Canberra->Australia    +61 (402) 700 883