Hello,
I am performing DTI experiments with b=1000 s/mm2 and 30 diffusion encoding
gradients. The scanner is GE 3T Signa HDx MR scanner. In the post-processing
stage, I ran 'bedpostx' using gradients from the scanner and finally
'probtrackx'. I noticed that the tracking results were not correct for standard
brain structures. While fixing this problem, I found out that my Nifti dataset
was not in radiological orientation. So, I changed it using the commands
fslswapdim input -x y z output
fslorient -swaporient output
But, I also had to change signs of 'y' values in the 'bvecs' (ie, y=-y) to get
correct tracking in 'probtrackx'. My questions are: what is the proper way to
verify the directions in 'bvecs' ? Does 'bedpostx' and 'probtrackx' assume
certain convention in 'bvecs'? Your suggestions are welcome.
Thanks in advance.
Shantanu
|