Hi Nele
For every volume in your data file, you need one corresponding bval entry and one bvec entry. A bval entry is a scalar, while a bvec entry is a 3x1 unit vector. So in your case you need a bvals file with 16 numbers and a bvecs file with 16 vectors. Have a look at
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT/UserGuide#DTIFIT
to make sure that you get the expected format right.
Hope this helps
Stam
On 15 Jan 2014, at 10:39, Nele <[log in to unmask]>
wrote:
> Hi,
>
> I want to analyze DTI-images using FSL. I've received a dataset of images ( in nii.gz format, recorded in 3 different hospitals) and one text-file with bvecs and one with bvals.
>
> I had 16 files per participant that I merged with dcm2niigui 3D>4D. Afterwards I corrected for Eddy currents and I made a brain mask. When I inputted everything (ecc corrected image, brain mask and text-files) in DTIFIT I got the error: 'data bvecs and bvals do not contain the same number of entries'. Indeed, I had 16 bvals (the same number as images) and 48 bvecs (3 times as much). If I understand it correctly, I should also have 16 bvecs (and I assumed maybe that there was one line of bvecs per hospital)!? Is this a correct assumption?
> Consequently, I tried the analysis with 16 bvecs (all 3 options) but this resulted in the error: 'terminate called after throwing an instance of ‘NEWMAT::IndexException’.
>
> I'm new to FSL so I don't understand everything yet, can anybody tell me what is going wrong?
>
> Thank you very much!
>
> Nele
|