[FieldTrip] mesh realignment

[Dr N.E. Adams]

Dear Jan-Mathijs

Ah, thank you, I believe this question has been answered for me, and it 
is really useful to know about ft_transform_geometry - that script had 
passed me by. I have now preprocessed my data in SPM and was looking to 
be able to look up atlas coordinates for the source-reconstruction step. 
The spatial model for the inversion is based on a mesh of the 
(coregistered) individual's brain shape, but an normalized version of 
this mesh acts as a look-up table for coordinates. My atlas needed to be 
registered to this normalized brain for me to know I was looking in the 
right area of the brain for specific brain regions based on the atlas 
label. All sorted now.

Thanks and kind regards
Natalie


On 2020-04-14 07:35, Schoffelen, J.M. (Jan Mathijs) wrote:
> Dear Natalie,
> 
> I am not sure whether I understand the reason why you would want this.
> Could you explain why you think you’d need to ‘align’ the meshes in 3D
> space?
> 
> Typically, when doing a cortical mesh based source reconstruction,
> while at the same time using information from an (anatomical) that’s
> defined on a cortical mesh, the game would be to have the topologies
> of the meshes aligned, and not the exact position of the geometric
> objects. In other words, the goal (and recommended approach) is to use
> surface based registration techniques (and possibly re-tessellation)
> to have a one-to-one correspondence of the vertices in both meshes.
> 
> If, however, you insist on doing a 3D alignment using icp (and
> assuming that it worked), you can apply the affine matrix to the mesh
> using the ft_transform_geometry function.
> 
> I would not try to do an intermediate interpolation to 3D space.
> 
> Best wishes,
> Jan-Mathijs
> 
> 
>> On 13 Apr 2020, at 10:07, Dr N.E. Adams <nea26 at cam.ac.uk> wrote:
>> 
>> Dear Fieldtrip Community
>> 
>> I would like to align two brain meshes and I can't figure out how.
>> Specifically, I would like to align a MEG-coregistered segmented brain 
>> mesh that has been normalized (mesh_brain) to an atlas cortical mesh 
>> (mesh_atlas). The two meshes, in this example, are about 20mm off 
>> alignment with each other in the rostral-caudal dimensnion in the 
>> attached images and I'd like to correct for this sort of alignment 
>> issue.
>> 
>> Can I do this with ft_volumerealign()? If I can, I can't figure out 
>> the settings.
>> 
>> Attempted workaround:
>> I tried using icp() to get the affine to tranform one to the other, 
>> but I do not know how to then apply this affine to meshes rather than 
>> volumes, nor if this has worked. I tried converting the mesh to a 
>> volume, but this was only partially successful and I'm unclear how to 
>> convert it back. Thinking this was a rabbit-hole I thought there must 
>> be an easier solution. Anyone know?
>> 
>> Thank you in advance for you help
>> Natalie 
>> Adams<mesh_atlas.JPG><mesh_brain.JPG>_______________________________________________
>> fieldtrip mailing list
>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>> https://doi.org/10.1371/journal.pcbi.1002202
> 
> 
> _______________________________________________
> fieldtrip mailing list
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> https://doi.org/10.1371/journal.pcbi.1002202