Dear Saad,

thanks for your quick response. I still believe, that my voxel size has dto do with the problem, because after reslicing images at 2x2x2mm3 voxel size probtrack runs without any problem. Could it be a problem of memory (I already tried to set swap space to 7 GB, but this did not help)?

And some other question: is there a way to simply reslice images with another voxel size using fsl (i.e. reslizing from 1x1x2mm to 2x2x2mm?).

Best wishes, Götz

Saad Jbabdi schrieb:

[log in to unmask]" type="cite">

Hi - I don't know why you've got this error, but it is unlikely to be due to
the voxel size. The tractography works fine on anisotropic voxels, although
I wouldn't recommand it because one direction (z here) is averaged over a
longer distance, which could be misleading in the process of estimating the
pdf on diffusion directions. 

Saad.

On Wed, 24 Jan 2007 11:48:40 +0000, GÃ¶tz Thomalla
<[log in to unmask]> wrote:

Hi,

i recently reported a porblem with ProbTrack, and I wonder, if this might
result from the resolution of my images (voxel size 1x1x2mm). Is this
possible, that either the anisotropic voxel size, or the larger amount of
data in these images (compared to 2x2x2 images, where tracking works quite
well) might cause the failure of ProbTrack? And if so, might an increase in
swapfile or some other modification help, to get ProbTrack running on
1x1x2mm data?

I hope someone can help me with this,

see you, Götz Thomalla
 =======================================================================

-- 
Dr. med. Götz Thomalla
Klinik und Poliklinik für Neurologie
Universitätsklinikum Hamburg-Eppendorf
Martinistr. 52
20246 Hamburg
Tel. 040-42803-9901
Fax 040-42803-9955