FoVs are different

1

Hi MRtrix users/experts,

We’re working on a dataset that has acquired two dMRI sequences in two opposing phase-encoding directions (LR and RL). For some of our subjects, we noticed a slight difference in FoV, which disrupts preprocessing. The solution we had thought of so far:

  1. Convert both files into .mif (with -fslgrad bvec bval, of course)
  2. Extract the “transform” of the LR image using mrinfo and save it as transform.txt
  3. mrtransform seq_RL -linear transform.txt -template seq_LR seq_RL_transformed to align the transform of seq_RL to seq_LR

We’d greatly appreciate any comments on whether this approach is valid or not.

Best,
Amir