Dear Anderson,


Thank you for your response.


Misaligment between functional images and anatomical/MNI template may result, using FNIRT, from the non brain extracted template, isn't it?



Best regards,


Christophe


De : FSL - FMRIB's Software Library <[log in to unmask]> de la part de Anderson M. Winkler <[log in to unmask]>
Envoyé : lundi 30 avril 2018 15:44
À : [log in to unmask]
Objet : Re: [FSL] FNIRT MELODIC group-analysis
 
Hi Christophe,

Registrations that don't look good are not normal. Small misalignments unfortunately always happen, but we don't tend to consider poor registrations as normal...

All the best,

Anderson


On 25 April 2018 at 15:24, HABAS Christophe <[log in to unmask]> wrote:

Dear Anderson,


Thank you!


Just another question:


when I perform FNIRT coregistration, registration between functional images and anatomical/ MNI template looks not very good...is it normal ?



many thanks.


Best regards,


Christophe


De : FSL - FMRIB's Software Library <[log in to unmask]> de la part de Anderson M. Winkler <[log in to unmask]>
Envoyé : lundi 23 avril 2018 05:17

À : [log in to unmask]
Objet : Re: [FSL] FNIRT MELODIC group-analysis
 
Hi Christophe,

If you have N subjects, indicate that in the tab "Data". Then in the "Registration" tab, when you click on "Select main structural images", you can indicate the N structural images for the respective subjects.

All the best,

Anderson


On 16 April 2018 at 06:32, HABAS Christophe <[log in to unmask]> wrote:

Dear Anderson,


Thank you.


But, for MELODIC group-analysis, what can be the Main Structural Image required for the non linear registration?


Again thanks.


With best regards,

Christophe


De : FSL - FMRIB's Software Library <[log in to unmask]> de la part de Anderson M. Winkler <[log in to unmask]>
Envoyé : lundi 16 avril 2018 02:01
À : [log in to unmask]
Objet : Re: [FSL] FNIRT MELODIC group-analysis
 
Hi Christophe,

Yes, use the option "Nonlinear" in the Registration tab.

All the best,

Anderson


On 12 April 2018 at 06:48, Christophe HABAS <[log in to unmask]> wrote:
Dear Experts,

Is it to possible to perform FNIRT registration with MELODIC for multisubject analysis?

Thank you.

Best,

Christophe