Dear Experts
Hi
I'm trying to simulate EEG data. Actually, I'm going to assign a focal source with arbitrary sinusoidal signal to an arbitrary region of head volume and Obtain the signals recorded at the electrodes on the scalp with an arbitrary electrodes' structure (cap).
I've almost followed all discussions about this subject like:
but I've couldn't do it. I've completely confused on doing process steps. I've done the steps as follow:
1- create protocol and subject (use default anatomy and use one channel file per acquisitions )
2- Set a default EEG cap
3-Edit the EEG cap to fit in on the subject’s head
4-Compute a forward model (whole cerebral volume)
5-Define some source activity, with the process "Simulate > Simulate generic signal.
6-Create a scout for which you want to set this activity (you can use the menu “Scout > New: Coordinates” to create a new ROI based on MNI coordinates)
7-Associate the signal with the scout and use this to simulate EEG or MEG recordings, with process “Simulate > Simulate recordings from scouts” (select the simulated signal in Process1, and select the scout in the process options)
Are these process steps right? Do I must to do additional steps that I left it?
That seems about right. You need an anatomy with registered EEG electrodes. With that a forward model that relates source activity in the brain to the electrodes. You could do this with a forward model with just the cortical surface as sources usually.
If you did those steps you can check what kind of simulated EEG signals come from that and see the topological plots and the frequency plots to see if that 's more or less what you expect.
How can I introduce an anatomy with registered EEG electrodes? could you please tell me which step?
I have another question, please. Can I compute leadfield matrix by BS and export that to the database and import the computed leadfield for another protocol?
Can you tell me what you exactly mean with this? Do you have an individual anatomy that you would like to use for your simulations?
You can compute a leadfield matrix and export that to for example matlab and import it again. Though the leadfield matrix is very specific to all the variables (anatomy, EEG locations, segmentation, BEM or 3 shell etc.). Why would you like to do that?
At first, I want to create a sinusoidal focal source in default BS anatomy and simulate the achieved signals from it in electrodes.
I just want to move the focal source in several regions of whole head volume with the same anatomy and electrode positions (cap). Since computing leadfield is a Time-consuming process for each scenario and protocol, I want to compute the leadfield matrix once.
Hi Francois,
At first, thank you for your valuable addition.
I tried to simulate EEG signals from dipoles in a new protocol without preprocesses that were done in "TutorialIntroduction". So when I wanted to do this step " Add process Simulate > Simulate recordings from dipoles" I couldn't find it. Could you please tell me is it necessary to do the preprocess in "TutorialIntroduction" or no?
but when I'm trying to assign the simulated signal "Simulated X (3*600)", I face with this error:
I know that it is possibly related to constrained scout, but it isn't mentioned that how I chang constrained scout to unconstrained scout.
how can I fix this error?
No, you can run this from anywhere, but you need to update Brainstorm first.
If you are using an older version of Matlab, maybe there is some syntax in this new process that is not recognized. Do you see any error/ message in the Matlab command window when you click on the button "Run" of the Process1 tab?
but when I'm trying to assign the simulated signal "Simulated X (3*600)", I face with this error:
This error message also indicates that you are not running the latest version of Brainstorm. The error message is slightly different now. Please start by updating Brainstorm.
