Hi Saad,
The RGB with the V1 looks way off. For eg, the corpus callosum does not indicate the correct colour. In general the RGB overlay of the V1 to the FA map does not have "strong" colours overlayed onto it. The S0 too, is not a T2 image, and it should be. The FA map in general just looks extremely noisy. I have compared this data on DTI studio and it looks much different (and cleaner) with DTI studio.
When you tried this at your end, can you please send me a screen shot of what the FA map along with the added V1 looks like?
Thank you, Saad.
David
On 6/12/07, Saad Jbabdi <[log in to unmask]> wrote:
Hi - I have tried dtifit on your data, and everything seems OK to me.
The only problem is one slice (Z=15) which appears to have a huge signal
drop at the 4th gradient direction. Everything else is OK (FA map, V1,
S0).
Also, the data are upsampled to 256x256, which is not recommended.
When you say your FA map is "quite bad", what do you mean exactely?
Saad.
> Dear FSLers,
>
> I am trying to run dtifit and the FA map is quite bad. I have uploaded
> my 4D eddy corrected niftii file to the FSL server (as well as the brain
> mask, bvals and bvecs) and the reference number is: 278630
>
>
> What is also strange is that one of the outputs from dtifit --_S0 --
> ie the raw T2 signal with no diffusion weighting, is NOT a T2 signal.
> In addition, I am able to see my 2 reference scans (B0) in fslview which
> are clearly T2 images, but still, the output from dtifit is NOT a T2
> image, and to my knowledge should be.
>
> Can someone please help! Our MRI physicist does not know what could be
> going on here with the 4D file and/or bvecs/bvals files which dtifit is
> utilizing.
>
> Thank you in advance.
>
> David Kideckel, PhD Candidate
