Hi Shie,
You won't get good results trying to register an ICA map to a T1
image. You will need to calculate a non-linear registration between
the two T1 spaces (i.e. from the GenR130 template to the MNI152
template) using flirt+fnirt, then apply that registration to your IC
maps.
Cheers,
Paul
On 04/03/2019, [log in to unmask] <[log in to unmask]> wrote:
> Dear FSL experts,
>
> I'm working on resting-state data and want to compare the results of a group
> ICA with an ICA networks template (using fslcc). Our participants are 8-12
> years old, and therefore we used pediatric template for registration.
>
> When I try to register the network templates to the our standard space the
> results I get are a bit off. I tried several different network templates
> (such as the pediatrics template here: https://www.nitrc.org/projects/genr/
> ; Yeo's 7 networks; and Beckmann's ICA networks), but the registration is
> always a bit-off (when I open both images in fsleyes mid-line is not
> aligned, or one image is a bit 'smaller' although the dimensions of both
> images are the same according to the fslinfo).
>
> The same thing happened when I tried to register the group-melodic output to
> the MNI152 (only for comparison purposes).
>
> I have used the following commands:
>
> Network-template to standard:
> flirt -in network_template.nii.gz -ref standard_image.nii.gz -out
> networks_reg.nii.gz -applyxfm
>
> (with and without the options: '-interp nearestneibghbour' '-usesqform' and-
> '-applyisoxfm' [instead of 'applyxfm']).
>
> Using Flirt gui I also tried to do translation only/ rigid body (both with
> the option 'not aligned, same orientation' and 'incorrectly orient') and got
> a file with only one component that didn't match any of the original
> components.
>
> ICA results to MNI152:
> flirt -in melodic_ICs.nii.gz -ref MNI152.nii.gz -out newICs_reg.nii.gz
> -applyxfm
>
> I searched the forum for solutions but couldn't find any.... Would
> appreciate any suggestions.
>
> Another question I have is about using fslsplit (and later- fslmerge) on the
> melodicIC.nii.gz (so I can use only the chosen components in the dual
> regression stage). What is the right way to do that (using -t/ x/ y/ or
> -z)?
>
>
> Thank you!
> Shie Rinat
> University of British Columbia
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|