Fixing Left-Right Reversal

, ,
1

Hi,everyone,
I recently tried running fMRIprep after performing left-right hemisphere reversal, but I’m unsure about the correct way to accomplish this. I attempted two approaches:

  1. Following the method on FreeSurfer’s LeftRightReversal page, I used mri_convert --in_orientation to modify the header orientation (for both structural and functional images) before running fMRIprep.
  2. Alternatively, I used FSL’s fslswapdim to flip the matrix along the left-right axis before running fMRIprep.

However, the resulting BOLD signals from these two methods are not identical. Specifically, which of the suggested methods (mri_convert vs. fslswapdim) will accomplish my goals? Or another suitable method.

I’d appreciate any insights or suggestions!

2

Our usual practice is to use afni 3dresample to put the images into LPI (right-handed) orientation after running fMRIprep, not before. We use LPI since it’s useful to have a default and LPI is handled consistently by the software we use, but if the NIfTI headers are correct you should be able view an image however you prefer (left-on-left or left-on-right) without having to change the actual image file.