EEG generation from Kuramoto sheet

Hi Mike,

hope & trust you are doing well.

I am intending on using a sheet Kuramoto model to produce BOLD/EEG data.

For the BOLD data i’m going to use the HRF/Balloon-Windkessel function.

For the EEG i’m not sure how to go about it. Convolution via a pre-computed IRF has been suggested but i’m not sure to which I can refer exactly. If it makes a difference, the intent is to simulate particularly sheet in motor cortex.

Do you have any suggestions and would your ‘Generate and visualize data’ Udemy course help to this end?

Any info justifying such EEG data generation would also be of interest!

Thanks in advance and stay safe,
Peter

Hi Peter. You can use a 1D Kuramoto model to generate non-sinusoidal oscillations that are your EEG signals. Adding some random 1/f noise (e.g., pink noise) will give it a pretty realistic texture.

Or are you using the 2D Kuramoto model to specify the spatial correlations only? In that case, you can make each node generate EEG as 1/f noise + some signal, like a sine wave, narrowband filtered noise, or a 1D Kuramoto model.

Thanks for the quick response!

It’s a 2D sheet intended to reproduce spatial patterning phenomena (from Heitmann et al. 2012).

Is there any biological justification gained by going with the 1D Kuramoto model over a sine wave? Or any special considerations for simulating motor cortex/beta activity?

I’m also curious as to why you ask ‘spatial correlations only’?

Well, I was thinking that a 1D model produces a time series at a single point (one voxel or dipole), whereas a 2D model produces a single numerical value for each of many spatial points. EEG is a dynamic signal, so people generally simulate a time series rather than just one point. With a 2D Kuramoto sheet, you might still want to generate a time series for each vertex, then scale that time series by the spatial value.

To be honest, I’m kindof just making this stuff up. I’ve never used a Kuramoto sheet to simulate data, so I’m trying to be helpful by thinking about how I’d go about it. I hope it’s more helpful than confusing :stuck_out_tongue_winking_eye:

You are preaching Mike, this choir has also never done this before!

I follow your thinking and it sounds sound to me…

Appreciate the borrowing of your brain for my task! Unsurprisingly helpful imo :grin:

I’ll hopefully post an in-progress or complete update along the road! Cheers