Source Estimation on EEG data

[ashenfernando1]

@MertBuyulu, documented below are some of the information I have regarding this issue.

What we have:

• raw EEG data which captures voltages on the scalp

What we want (ideally):

• deep brain sources from which these voltages are emanating from

We'll be using the MNE library (https://mne.tools/stable/index.html) for our purposes. There will be many approximations done because we actually need fMRI data of individual subjects to produce unique source volumes but unfortunately that's not feasible in our scope.

I will share a notebook and some of the already-generated files needed to produce the visualizations through Teams.

Another aspect of EEG data is that there are a lot of artifacts, noisy data that obscures 'signal'. One of the ways to remove this noise is through what is called Independent Component Analysis (ICA) and MNE is an excellent tool in this aspect as well. What would be ideal is if we could automate this process, and there are methods already in place within the library to do so, but would need more work to be implemented.

Since this is a lot, ping me any time through the week as you go through this stuff.

Potentially useful links:
https://jasmainak.github.io/mne-workshop-brown/evoked_to_stc/stc
https://mne.tools/stable/auto_tutorials/preprocessing/40_artifact_correction_ica.html#tut-artifact-ica