Dear FSL users and FMRIB group,
I am performing research on Diffusion Weighted images Rats for which I use FSL. I want to use TBSS to analyse my created FA maps (from DTIFIT, I have two groups of animals), but have run into problems when using tbss_2_reg. To better fit my images (resolution 0.25x0.188x0.5mm, 128x128x26 pixels), I have modified some parameters in the configuration files below, I also changed some templates (they are now images from my population). Running tbss_1_preproc works fine.
When running tbss_2_reg, it starts but does not fulfill the registration (it creates empty _to_target_warp.msf files and _to_target.mat of 180 byte). To go around the problem, I tried to do the registration using FNIRT (on original FA images before tbss_1_preproc), from which I get the following message "Incompatible combination of --subsamp, --warpres and voxel size in x-direction". Clearly, some of my modifications does not work. Could anyone please explain to me what the requirements for the parameters are?
This is how I called FNIRT
fnirt --ref=FMRIB58_FA_1mm.nii.gz --in=FDT_4plus40_FA_FA.nii.gz --config=FA_2_FMRIB58_1mm.cnf -v
Where the --ref image is an FA image of same dimensions and pixel resolution as the --in image (resolution 0.25x0.188x0.5, dim 128x128x26).
The changes I performed in the .cnf file are
--infwhm = 0.8,0.8,0.2,0.2
--reffwhm = 0.8,0.8,0.2,0.2
--warpres = 1,1,1
I have previously run TOPUP on the same images (before DTIFIT) with --fwhm=0.8 and --warpres=1, which worked fine.
Thank you in advance!
Regards, Sandra
|