Hi,
thanks for the reply!
Initially, I registered the mask to diffusion space, but the results didnt look so good. I'm masking a small structure (hypothalamus), and for my analysis, I want to exclude e.g. CSF voxels. But after re-binarizing the mask after registration (flirt) with different thresholds,
the mask always includes too much CSF but excludes other important parts of the hypothalamus, and looks very different across subjects.
I cannot initially register the T1 to DWI space, since I have to create the mask in AC-PC orientation.
Therefore I thought it might be reasonable to bring the DWI into AC-PC orientation instead. This way, I'm getting rid of one registration step that changes my mask.
What I did now:
- run epi_reg on the first b0 image of the eddy corrected data
- downsample my T1 to 2x2x2
- apply the epi_reg transform to eddy_corrected_data using the downsampled T1 as reference.
- bvec rotation using the link you provided
The T1 and eddy_corrected data now have the same dimensions, V1 and FA look correct and my mask is better aligned.
Does that sound like a reasonable procedure to you?
Thank you for your input!
Cheers
Melanie
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|