I have 64 channel EEG data that I would like to analyze in the source space. The electrode locations were not digitized nor were subject level structural scans obtained. Is it possible to perform this analysis in Brainstorm (using coarse assumptions)?
If you don't have the individual anatomy of your subejcts, you can use the anatomy templates in Brainstorm:
https://neuroimage.usc.edu/brainstorm/Tutorials/DefaultAnatomy
If you don't have individual electrode positions, you may use default positions for the corresponding template. Add them to an existing channel file with the menu "Add EEG positions".
You should still configure your patients with the option "Default channel file: NO" if you have multiple acquisition runs per subject, or "Yes, use one channel file per subject" if you import only one file per subject. You need to estimate the noise covariance and sources separately for each subject.
These are of course not optimal solutions for estimating sources, however it can still you a global idea of the origin of the brain activity, at least at the level of the lobe. Just keep in mind in the exploration of your results that the localizations you obtain are coarse approximations.
https://neuroimage.usc.edu/forums/t/easycap-64-channel-location-setup/1027/2
Hi Francois,
I've been puzzled by a problem for a long time. I want to compute sources ,but i don't have the channel location. Now, i use default EEG cap, am i right? If it's wrong, what should i do?
Thanks for your help
luzhou
If you don't have individual electrode positions, you may use default positions. Add them to an existing channel file with the menu "Add EEG positions".
If you are using them on an individual brain, you need to align and project them on the head surface.
https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy#Prepare_the_channel_file
Hi Francois,
I want to replace the electrode positions in an existing channel file with default electrode positions. I'm wondering if the following steps are correct?
- To remove current electrode positions: 'Digitized head points'> 'Remove all points'
- "Add EEG positions" and select default positions (e.g., ICBM152>EGI>GSN HydroCel 128 E1)
- Check positions using 'MRI registration'> 'Check'
Thank you so much in advance,
Jane
Hi Jane,
You can skip the first step, this is to remove the additional digitized head points, not the positions of the electrodes (ie. the extra hundred points we recommend to acquire for MEG registration). You probably don't have any, so it doesn't do any harm, but it's useless.
If you use standard electrode positions on the head of the subject for which you have the anatomical MR, you may need to manually adjust the electrodes on the scalp (right-click on the channel file > MRI registration > Edit)
Hi Francois,
Apologies, I should have provided more information about my situation. I import data preprocessed from EEGLAB, and uses a default anatomy for my protocol because I do not have MRI data for my subjects. When importing into Brainstorm the existing electrode positions on data channel file are quite askewed from the default anatomy, and while I can manually adjust the electrodes it will take a lot of time because I use one channel file per acquisition run. Since I do not have individual MRIs for subjects anyway I am intending on replacing the current electrode positions in the channel files with default positions. I tried initially to 'Add EEG positions' straightaway to my channel file, but when I check the positions using MRI registration, I see two caps on the MRI. That's why I tried removing the digitized head points, which looks like it works because the resulting MRI registration looks good. But I'm concerned as to whether I'm doing it the right way. Do you think my approach is okay / problematic?
WIth much thanks,
Jane
I see, what you do makes sense then.
EEGLAB probably adds default non-realistic positions to your dataset for display purposes (2D topography plots). When importing the dataset from EEGLAB, these positions are imported as 3D electrode positions AND copied as additional head points for allowing automatic registration (in case they were properly digitized, as in this example: https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy#Register_electrodes_with_MRI)
When replacing the electrodes positions with the template positions in Brainstorm, the additional head points copy is not removed, so you needed to remove it manually.
This was not very intuitive, so I changed this behavior: now when adding electrodes positions from a template, it removes these additional headpoints automatically:
https://github.com/brainstorm-tools/brainstorm3/commit/e13f0ba2bf3a16b635d8ad82ab8e1f479bdfa50f#diff-647bcb5fd5b51b8fc3bc295dbd7a2ac0
Update Brainstorm to get this last modification.
Thanks for bringing up this issue.
Thank you so much Francois! I really appreciate your prompt replies and action!
Cheers,
Jane