[Homer-users] Atlas viewer subject specific

[Theresa Sukal Moulton]

Hi Homer and AtlasViewer users,

I’m hoping that someone may have some ideas for me in problem solving participant-specific registration of probes in AtlasViewer. Some steps have worked beautifully, but I’m getting stuck on others. I’m listing what I’ve done here to check if I’ve done something wrong but hopefully to also respond to an earlier post asking for the step-by-step registration instructions.

1 — Run recon_all in Freesurfer for a participant’s MRI. Then I used the mri_convert commands from the command line to change T1.mgz, aseg.mgz, brain.mgz, and wm.mgz to be .nii files. Do a check of the segmentations with Freeview.

2 – Add a copy of the .sd file for my probe to folder where the Freesurfer output is, as well as a digpts.txt for the digitized locations of the scalp landmarks plus optode locations.

3 - Launch AtlasViewerGUI from that folder.

4 - Select File -> Convert FS files to Viewer. The images are really nice after this step, and I can see also my digpts.

5 — Click Tool -> Find Ref pts. Here, I identify manually Iz,Nz,Cz,A1, and A2 and then click the save ref pts button.

However, when I go back to the main AtlasViewerGUI window, the points that I’ve just selected (viewed by “show ref pts labels” radio button) are not oriented correctly on the head model (see attached image). The points are selected and saved in a different order than AtlasViewer is expecting them, but that is solved by a function already written. I end up with an identity matrix for the refpts2vol.txt – perhaps that’s correct because they are defined in the individuals brain volume space… but then how is the head surface being displayed in a different space within the AtlasViewerGUI window?

If this alignment were working properly, I believe the next steps are to calculate the 10-20 points, and then register the digpts to the anatomy, followed by register probe to surface. I can manually help the program get through these steps but not surprisingly the results are clearly incorrect.

Finally, I receive an error if I attempt to project channels or optodes to the cortex that I do not have cortical anatomical labels for this anatomy. Is there a way I can use my Freesurfer output to create the required Labelssurf.mat file (using mris_label2annot or some other function?), or can I bypass the need for labels and just get the coordinates of the projected channels or optodes?

Many thanks for any input you may have on my challenges. We’ve done a lot of registration to the Colin Atlas and use of the probe placement variation tool, and are very interested to take it to the next level with the participant-specific registration.

 

Kind regards,

Theresa Moulton







-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20150828/95b4bb81/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-1.pdf
Type: application/pdf
Size: 75869 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20150828/95b4bb81/attachment.pdf 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20150828/95b4bb81/attachment-0001.html