ftadel
March 6, 2019, 11:17pm
1
Related documentation: https://github.com/bids-standard/bids-specification/blob/master/src/99-appendices/08-coordinate-systems.md
I’m trying to find out how to name correctly my ieeg files:
sub-01_ses-01_space-_electrodes.tsv
sub-01_ses-01_space-_coordsystem.json
What is the recommended space label to refer to the head coordinates (in subject space, not normalized) coming from the voxel-to-world transformation sform coming from a .nii file immediately after converting from DICOM?
“head”, “world”, “ras”, “orig”, “subject”?
Is it OK to specify only MNI152, or should we always specify the full template name? (eg. MNI152NLin6)
If the latter: does anybody know what to indicate when using an affine SPM12 registration (spm_maff8 )?
1 Like
ftadel:
What is the recommended space label to refer to the head coordinates (in subject space, not normalized) coming from the voxel-to-world transformation sform coming from a .nii file immediately after converting from DICOM?
“head”, “world”, “ras”, “orig”, “subject”?
The orig
value that used to serve this purpose was removed in the last minute before releasing 1.2.0 . The only option I can see within the current spec is setting space
to Other
as described https://bids-specification.readthedocs.io/en/stable/99-appendices/08-coordinate-systems.html#coordinate-systems-applicable-to-meg-eeg-and-ieeg . Perhaps others can chime in.
This is a controlled vocabulary so you need to pick full template name.
AFAIK SPM uses different templates for different procedures. Perhaps @Guillaume can shed some light here.
ftadel
March 7, 2019, 7:49am
3
The orig
value that used to serve this purpose was removed in the last minute before releasing 1.2.0 .
I’m sorry I missed this deadline, I would have insisted a bit more on these two questions. Both are related to this example (naming the two pairs of electrodes.tsv/coordsystem.json):
https://github.com/bids-standard/bids-examples/tree/bep010_ieeg/ieeg_epilepsy/sub-01/ses-postimp/ieeg
The only option I can see within the current spec is setting space
to Other
This is going to be the standard case for positioning intracranial electrodes, we should have something a bit more informative.
1 Like
I previously used space-orig. For my ieeg-example data the MRI was in ACPC and I could use that ACPC label.
For my own data, after the -orig label was removed I labeled “iEEGCoordinateSystem” as Other and described in iEEGCoordinateSystemDescription.
1 Like