Hello,
I have a question about the application of cost function masking in a sample of tumour patients. I would like to run the MELODIC tool to analyze my resting-state fMRI data for group-analysis. We have already created the single-subjects lesion masks. We went through the theory of cost function masking to deal with brain tumour patients as it seems to be one of the best strategies to solve normalization problems. We still have some questions regarding the application of it.
Question 1:
Is there any chance to include the lesion masks (or the structural image masked) in the FSL MELODIC GUI at the registration step? We weren't able to do it.
Question 2:
Related to the above problem, we thought to do the registration step out of the GUI, first coregistering the rs-fMRI data to the T1 image with the following command line:
flirt -in "$patient_name"_rsfmri_feat_bet.nii.gz -ref "$patient_name"_T13Dax_bet.nii.gz -out "$patient_name"_rsfmri_bet_to_T1.nii.gz -omat "$patient_name"_rsfmri_to_T1.mat -bins 256 -cost normmi -searchrx -90 90 -searchry -90 90 -searchrz -90 90 -dof 12 -interp trilinear
The output is a file too big to be used by MELODIC (21 GB), because the rs-fMRI was resampled to the T1 voxel size. So we added "-noresample" to the previous command but it didn't work.
Is there any strategy to apply the cost function masking to the T1 before registering to the MNI?
Could it be a correct alternative to use only FLIRT and FNIRT without cost function masking and check the quality of registration for larger lesions?
Many thanks in advance
Best,
Beatrice
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|