Importing Polhemus locations (when polhemus recording is not driven by brainstorm)

Discussions
1

Hello,

I am a new brainstorm user, still making my way around understanding how it works. I’ve imported eeg data and a channel position file (129 channel spf file converted to txt attached - GSN-HydroCel-129new_sfp.txt).
Before recording, I digitized several head points for each subject (unfortunately only a few) using the polhemus Patriot system.
However I did this independantly of brainstorm (did not use Brainstorm to drive it). I am having trouble figuring out how to import the polhemus locations and impose
them on (e.g., change) the channel file I already have (only a few channels will be changed based on the polhemus recording since I did not collect many, the rest should stay as they are). The polhemus file (attached)
does not use the same coordinate system as the spf file with the standard channel location of the system I used (EGI 128 channel), and of course it is relative to the digitizer (I need to re-reference to VREF?).

Thanks so much for your time!
Jacqueline

2

Hi Jacqueline,

  1. You are using a default set of positions for this EEG cap, right?
    You can use the default positions available in Brainstorm instead (menu Use default EEG cap > Colin27 > GSN Hydrocel 128).

  2. Your two files are in different coordinate systems and you don’t have any anatomical references to align them together, or to align them with the anatomy in Brainstorm. You will not be able to register them, there is too much missing information.
    To get all your files registered correctly, you need common references. You should always digitize the three referential points we use in Brainstorm: NAS, LPA, RPA. The coordinates of those points should be available in all the files with 3D positions, Brainstorm would convert them automatically to the appropriate coordinate system.
    http://neuroimage.usc.edu/brainstorm/CoordinateSystems

  3. An easy solution for getting correct Polhemus files is to use the Brainstorm digitizer:
    http://neuroimage.usc.edu/brainstorm/Tutorials/TutDigitize

  4. You should use the menu “Add EEG positions” instead of “Import channel file”, as illustrated in the EEG/Epilepsy tutorial:
    http://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy
    Repeat twice this operation: once for the default positions, once for the positions you want to redefine.

  5. I would advise you to always digitize the positions of all the electrodes. It will be difficult to mix the default positions with realistic positions…

Cheers,
Francois

3

Thanks so much Francois for your speedy response! So you think I cannot use the locations I have collected at all? I’ve already got all my data, so I cannot rerun with the brainstorm digitizer unfortunately, but I will know this for next time.
I’ve looked up the equivalent for the 128 GSN channels for the NAS and LPA and RPA, and unfortunately I did not digitize all of these (only NAS), but I do have T3 (E45) and T4 (E108) digitized which are near the LPA and RPA.

Do you think a source reconstruction will be possible without the polhemus localizations (with just the default position file)?

Thank you again, I really appreciate your help,
Jacqueline

4

So you think I cannot use the locations I have collected at all?

Well, unless you have some additional data with which you can register the two files...
You need to have exactly the same points defined in all the files you want to process together.

Do you think a source reconstruction will be possible without the polhemus localizations (with just the default position file)?

Yes, no problem with this.
If you have electrodes that you know are in a different place, you can move them manually on the head surface (right-click on the channel file > MRI registration > Edit). Then read the tooltips of all the buttons for help.

Next time, digitize the position of all the electrodes, plus the three fiducials (Nasion, Left, Right). This will also allow you to warp the template anatomy if you don't have the individual MRI scans of your subjects:
http://neuroimage.usc.edu/brainstorm/Tutorials/TutWarping

5

Thank you for your help Francois :slight_smile: it’s been a pleasure