MEG-MRI coordinate system alignment

✉️ Mailing List Archive (read-only)
1

Hi,

According to the
manual<http://martinos.org/mne/manual/analyze.html#coordinate-frame-alignment>,
the MRI-MEG coordinate frame alignment tools included in mne_analyze
utilize the 3D digitizer (Polhemus) data acquired in the beginning of each
MEG/EEG session and the scalp surface triangulation. Specifically what
scalp surface triangulation is used here? Is this something generated
during FreeSurfer's recon-all stream, or is it the BEM mesh created with
mne_watershed_bem?

Ultimately, I am trying to understand how to solve this question: I
recently overwrote my output from mne_setup_source_space,
mne_watershed_bem, and mne_setup_forward_model. Do I need to re-register my
MEG/MRI data before proceeding to compute the forward solution?

Best wishes,

Vincent Rupp Jr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20130602/fd0aec9b/attachment.html

2

Hi,
   It uses whatever "subject-head.fif" surface lives in the Freesurfer
$SUBJECTS_DIR/$SUBJECT/bem/ directory. This is usually prepared by
mne_watershed_bem unless you created this using mne_surf2bem from
something else.

Also, the answer to your latter question depends on why you replaced the
surfaces from mne_watershed_bem...Assuming your previous registrations
were with a reasonable head surface, since the number of parameters for
the MEG head to MRI head transformation is small compared to the number of
digitizer points, I'd say it is not necessary to re-register.

HTH,
Regards,
Hari

3

hi Vincent,

if you look in:

https://github.com/mne-tools/mne-scripts/blob/master/sample-data/run_anatomy_tutorial.sh

you'll read:

# Make high resolution head surface
mkheadsurf -s ${SUBJECT}
mne_surf2bem --surf ${SUBJECTS_DIR}/${SUBJECT}/surf/lh.seghead --id 4
--check --fif ${SUBJECTS_DIR}/${SUBJECT}/bem/${SUBJECT}-head.fif
# if the previous command fails you can use the --force option.

if this step succeeds you'll have a high resolution scalp for
coregistration rather than the BEM surface.

HTH
Alex

4

Thanks very much for the help. Great info.

Best wishes,

Vincent Rupp Jr