Dear Mrtrix experts,
I am a first-time user of MRtrix, and I am having some problems in running dwibiascorrect on some of my subjects’ DWI images.
Preprocessing of my DWI images has not been done using MRtrix commands, but a colleague of mine has previously used FSL (therefore topup and eddy correct have been performed already), as we are also doing other analyses.
I have used the eddy_corrected_output.nii.gz that we got from FSL as the input for my structural connectivity analysis, and this is the list of commands that I have run so far:
-
mrconvert {i}_eddy_corrected_data.nii -fslgrad {i}_bvecs.txt bvals.txt ${i}_eddy_corrected_data.mif
-
dwibiascorrect ants {i}_eddy_corrected_data.mif {i}_eddy_corrected_data_unbiased.mif -bias bias.mif
-
dwi2mask -clean_scale 3 {i}_eddy_corrected_data_unbiased.mif {i}_dwi_mask.mif
-
dwi2response dhollander -voxels voxels.mif {i}_eddy_corrected_data_unbiased.mif wm.txt gm.txt csf.txt -mask {i}_dwi_mask.mif
-
dwi2fod msmt_csd {i}_eddy_corrected_data_unbiased.mif -mask {i}_dwi_mask.mif wm.txt wmfod.mif gm.txt gmfod.mif csf.txt csffod.mif
I have a cohort of 123 subjects to analyse, and everything that I have listed has run smoothly for all of them, except 6 participants.
Basically after the conversion in .mif format, when I ran dwibiascorrect I ended up with empty images (0 values).
I have tried skipping dwibiascorrect as it is not always recommended, but the same problem persisted (empty images). My data look fine, there are no artifacts.
I tried to follow the suggestions from this previous thread of comments:
but anything has changed.
Do you have any idea on why is that happening so far?
Thank you very much and have a lovely day!
Best wishes,
Michela