I imported the channel file containing the XYZ values of the electrode positions of Neuroscan Quikcap (128-ch NSL), which come with the CD provided by Neuroscan (Neuroscan 4.2, attachment1) to Brainstorm, because Neuroscan Quikcap (128-ch NSL) is not in the Brainstorm default EEG cap. The problem is that my electrode positions are located directly “on” the cortex of the default anatomy (attachment1 (128-ch), attachment2 (64-ch); 64-ch electrode cap is attached to enable comparison of the same number of electrodes as that with the Brainstorm-prepared Neuroscan cap), although the Neuroscan electrodes from Brainstorm are located on the scalp (attachment3 (64-ch)). I think the same XYZ coordinate system is used in Brainstorm and Neuroscan. Will there be any problems while conducting the source estimation in this situation? In cases where this may cause problems, please inform me of a suitable arrangement for my XYZ values.
The electrode positions you have imported, although they are in the right coordinate system, need to be adjusted to the default anatomy of Brainstorm. You can do so following the directions described in the following pages: http://neuroimage.usc.edu/brainstorm/Tutorials/TutImportRecordings#Channel_file (Menu: MRI Registration, Edit…).
Right click over your channel file, Choose MRI Registration, then Edit… from the pull-down menu. You will then be able to adjust the position and the scale of the electrode locations. Once this is getting really good, you may want to click on the icon that shows the electrode will be eventually projected onto their closest point on the scalp surface.
Once you are done with this, I would appreciate if you could please send us the corresponding channel file so that we make it part of the brainstorm distribution, and this will certainly help some of our users using the same EEG system with the same montage.
Could you send me by email all the electrodes positions you can find on the Neuroscan CD?
I will adapt them on Colin27 brain for you and include them in the Brainstorm distribution.
Thank you very much for your instructions and offer. I was able to adjust my electrode positions by following the directions you gave me. However, it is difficult for me to adjust all the electrodes manually, and when I click on the “project electrodes on scalp surface” icon, I find that the electrode positions are a little different on the left and right side. I will send the results and the corresponding channel file by email.
Glad it worked for you, Masahiro.
We’ll include this montage to the Brainstorm distribution: thank you for sharing!
We’ll also try to fix the asymmetries you are mentioning, if possible.
I have used spherical head model for source reconstruction. I am not sure if the electrode coodinates have also changed to match the sphere scalp or not?if so, how I can find the new coordinates of them?
The electrodes positions are not modified, the scalp is not “spherical”, nothing is changed.
The head is considered being a sphere, even though it is not.
In MEG, it does not matter much, as the magnetic fields flow through the living tissues almost as through the void.
In EEG, it does matter, because the way electrical currents flow in the head is influenced by the tissues that are crossed, the skull is a strong insulator. The spherical approximation is usually a bad choice for EEG.
[QUOTE=Francois;3314]The electrodes positions are not modified, the scalp is not “spherical”, nothing is changed.
The head is considered being a sphere, even though it is not.
In MEG, it does not matter much, as the magnetic fields flow through the living tissues almost as through the void.
In EEG, it does matter, because the way electrical currents flow in the head is influenced by the tissues that are crossed, the skull is a strong insulator. The spherical approximation is usually a bad choice for EEG.
Although, I must say that with a spherical head model, the electrode locations are projected onto the sphere that best fits either the scalp of the subject or the original electrode locations. This is also why you want to use a realistic head model with BEM when doing EEG: with this latter, the sensor location and the head geometry remain the ones from your subject. If you don’t have the individual MRI for your subjects, you can use the MRI template that is distributed with Brainstorm and use either the digitized electrode locations from your subjects (if available) or one of the EEG montages, also distributed with Brainstorm.
The tutorial pages describe these steps but let us know if you need more help with all that.