On 8/10/18 6:48 PM, Ben M wrote:

        External Email - Use Caution        

Hi Doug,

Thank you for your suggestions, I'll try them.
Could you please let me know if my code is actually doing what I intend it to do? Specifically, the idea is to perform functional-to-structural registration.

so first I did
$ bbregister --t2 --reg /data/bbregister/reg.dat --mov /data/mean_func.nii.gz --s 001
That is right. The only thing I would change is to use reg.lta instead of reg.dat. The lta file has more info in it and we're trying to get people to move to it.

I'm not sure if the --mov image should be the motion-corrected mean functional image or the skull-stripped anatomical image. I think if I want to do functional-to-structural registration it should actually be the anatomical image as --mov right? Or it doesn't matter which one I choose?
Yes, it matters. The mov should be the functional.

I then did
$ mri_vol2vol --mov /data/bet_structural.nii.gz --targ /data/brain.nii.gz --o /data/coregistered_structural.nii.gz --reg  /data/bbregister/reg.dat --inv –-nearest
 which I thinks applies the transformation matrix to the structural image, am I correct? In this case it would be a structural-to-functional registration I think.
Did you see the examples (run mri_vol2vol --help)
So, if I intend to perform functional-to-structural registration I should actually rather do 

$ bbregister --t2 --reg /data/bbregister/reg.dat --mov /data/bet_structural.nii.gz --s 001
$ mri_vol2vol --mov /data/mean_func.nii.gz --targ /data/brain.nii.gz --o /data/coregistered_functional.nii.gz --reg  /data/bbregister/reg.dat --inv –-nearest

Could you please confirm if this is correct?

Thank you very much,
Ben




Sometimes the FSL initialization will fail. If you are using FS version 
6, then run it without specifying --init-fsl and it will use mri_coreg, 
which is more robust. If your anatomical and functional were acquired at 
the same time, you might be able to get away with --init-header instead.

On 8/9/18 2:12 PM, Ben M wrote:
>
>         External Email - Use Caution
>
> Hi,
> I'm trying to coregister the strucural image to the mean functional 
> image using bbregister, so after reconall I did:
>
> bbregister --t2 --init-fsl --reg /data/bbregister/reg.dat --mov 
> /data/mean_func.nii.gz --s 001
>
> bet /data/structural.nii.gz /data/bet_structural.nii.gz
>
> #brain.nii.gz is the brain.mgz from reconall output (in freesurfer space)
> mri_vol2vol --mov /data/bet_structural.nii.gz --targ 
> /data/brain.nii.gz --o /data/coregistered_structural.nii.gz --reg 
> /data/bbregister/reg.dat --inv –-nearest
>
> But when I load the coregistered_structural.nii.gz and the 
> mean_func.nii.gz they are not aligned. I'm sure I'm doing something 
> wrong, could someone help me figure it out please?
>
> Thanks in advance,
> Ben



_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer