Hi
It sounds that some of the transformation or transformed masks you provide are not right. Can you check the following:
- Can you overlay the transformed masks on the diffusion data in fslview? Or when trying to do that you get an error?
- Presumably you haven't used the transformation of (1) in the second probtrackx call? You should not provide any transformation in the second call, since you have moved all your masks in diffusion space.
By the way, you can transform masks non-linearly in FSL or supply non-linear transforms directly into probtrackx. See http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/fdt/UserGuide#Registration_within_FDT
Cheers
Stam
On 30 Jul 2013, at 05:01, YM Huh wrote:
> Hi,
> I obtained whole brain connectivity matrix using probtrackX with two different methods.
>
> 1. use standard space seed mask(AAL) and apply standard to diffusion transform matrix automatically by probtraxkX (using linear transform).
>
> 2. transform standard space AAL seed masks into diffusion space (using non-linear transform in SPM) to make subject-specific seed masks, and put them on probtrackx.
>
> I have few questions to ask.
>
> 1) I found few seed masks that have no streamlines with any other regions on results obtained by second method.
> Is it possible to have no connections with any other regions and even with itself?
>
> 2) when I check transformed seed masks (second method) with MRIcron, It looks fine.
> But after I ran probtrackx, fiber tracts appear wrong places!
> I followed your instruction (set max step to 2 and run probtrackx), and fibers appeared wrong place, too. (part from seed mask)
>
> I tried second method to apply seed masks more accurately, but I can't figure out what is wrong with the method.
> I would appreciate any comments.
>
> YM Huh
>
|