Dear FSL experts,
i am reposting this query, can someone please help me in this ?
Thanks and best
Rabia 

On Sat, Jan 21, 2017 at 6:22 PM, Rabia Mazhar <[log in to unmask]> wrote:
Dear FSL experts,
Is there a command line way to do non linear registration in bedpostx ?
And is it recommended ? I am seeing some fibers outside the brain thats why I am concerned and want to do no linear registration to correct this.
At the moment the commands I am using for linear registration are as follows:

flirt -in nodif_brain_subject -ref /Volumes/MAC/DTI_subject/anat_brain.nii.gz -omat xfms/diff2str.mat -searchrx -90 90 -searchry -90 90 -searchrz -90 90 -dof 12 -cost mutualinfo

convert_xfm -omat xfms/str2diff.mat -inverse xfms/diff2str.mat

flirt -in /Volumes/MAC/subject/anat_brain.nii.gz -ref /usr/local/fsl/data/standard/MNI152_T1_2mm_brain.nii.gz -omat xfms/str2standard.mat -searchrx -90 90 -searchry -90 90 -searchrz -90 90 -dof 12 -cost corratio

convert_xfm -omat xfms/standard2str.mat -inverse xfms/str2standard.mat

convert_xfm -omat xfms/diff2standard.mat -concat xfms/str2standard.mat xfms/diff2str.mat

convert_xfm -omat xfms/standard2diff.mat -inverse xfms/diff2standard.mat



Any help is this regard will be extremely appreciated


Thanks and best

Rabia