[Neuroimaging] Reorienting bvecs in Dipy

[Vinit Srivastava]

Hi all,

I used FSL-FLIRT to coregister the diffusion-weighted imaging series to the
anatomical T1 template, which also outputted an affine transformation
matrix (as a .mat file).  I'd like to reorient the bvecs/b-matrix table to
compensate for any rotations and translations performed during the
coregistration process for downstream tractography application.

I understand that Dipy offers such a function to reorient bvecs to
compensate for motion.  However, I do not know how to use the outputted
affine matrix from FSL as an input for the affines argument in the function
reorient_bvecs(gtab, affines).  I get the following error:

'Number of affine transformations must match number of non-zero gradients'

I'm guessing that the argument affines requires 'n' number of affine
matrices?  If so, do I need to modify the FSL outputted affine 4x4 matrix
in some way?

Thanks for your help,

Vinny
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20171026/d11bea71/attachment.html>