Dear Hankee,
I can actually see both left and right segmentations in all
the correctly registered subjects when using fslview. However
with first_roi_slicesdir I see what you are seeing. This has
happened only once before here and I think it is to do with
unusual FOVs or intensities. It is a bug that we need to fix in
the creation of the webpage images - nothing to do with the
quality of the actual segmentations - but I will look into this more
and hopefully get a patch out soon. In the meantime I recommend
that you check your results by loading up *firstseg.nii.gz files on top
of the corresponding structural files.
As for the vertex analysis, it does not work because you have
not created your design matrix using the GLM GUI as the
documentation says you should. You have just created a
list of numbers in a text file which is not quite what is needed.
If you run the Glm GUI, use Higher Level mode and create
your design matrix that way then you should find that it
works fine. It did for me.
You should also not need to discard the subjects E05 and N06.
Just do the following:
imcp N06 N06_orig
fslswapdim N06 RL PA IS N06
and redo the run_first_all on this new N06 and it should work.
Also do the same for E05.
All the best,
Mark
On 6 Nov 2009, at 05:21, SUBSCRIBE FSL hankee wrote:
> Hi,Thank you for your useful advice.
> 1. I checked data registrations and definitely found that E05 and
> N06 are
> wrong. So I discarded this two subjects (E05 and N06 ),rewrite the
> design
> matrix file, and rerun concat_bvars command and first_utils command.
> But the
> same error still existed. I don't know why??
>
> 2. So when checked the data segmentations, I found that subcortex
> structure
> of some subjects only right hemisphere was labeled, the left
> hemiesphere was
> not. Only E07 and N04 's bilateral hemisphere were labeled. I don't
> know
> which one is right? how to solve this problem?
>
> 3. I assume that bilateral hemisphere labeled subjects were right.
> So I use
> E07 and N04 data, do one copy of E07 and N04 and rename them as rE07
> and
> rN04. So the sujects are E07,rE07,N04 and rN04 and do the Vertex
> Analysis
> (rewrite the design matrix file, and rerun concat_bvars command and
> first_utils
> command). To our disappointment, the same error still existed! We
> don't know
> how to use the FIRST tool?
>
> Pls help me. Thanks a lot!
>
