Clusters and ROI

Discussions
1

Good day!
I was reading about clustering in BS but I did not understand:

Is it possibly to set a ROI (scouts) using clusters of channels?

Can I perform PSD , connectivity or another analysis on clusters I made and how (I tried but can not find the way) ?

The cluster function is just for displaying purpose or actually is it a way to create channel based scouts?

Thanks in advance!

2

Clusters of electrodes are mostly there for display purpose.
They can be useful for overlaying one electrode over multiple files:
https://neuroimage.usc.edu/brainstorm/Tutorials/ChannelClusters

To get the cluster's time series, you have two options:

  • create a cluster, use the process Extract > Clusters time series, then compute what you need on the resulting file
  • Use a montage and the process "Standardize > Apply montage" to create a new signal in the files. Do not try to do any source modeling on the files you have modified in this way.
3

Thanks!!!

4

Hello @Francois,

I have created a few clusters containing signals from different electrodes contacts (each cluster = one brain region, eg hippocampus, frontal lobe and fusiform gyrus). I would like to calculate the connectivity matrix between these clusters. Is that possible in BrainStorm?

I am not sure that I understand the method you described above with the "Standardize > Apply montage". How is that related to the clusters? Could you detail this part a bit further?

Thank you for your help,

Anaïs

5

I'm not sure how an electrode can be attributed to the recording of the hippocampus.
You have to be careful with attribution of EEG electrode=brain region. EEG scalp potentials are very smooth, due to the skull and the diffusion of the currents in the head tissues. Some EEG features visible on most of the electrodes (eg. P300) can't be attributed to a single brain region.

I would like to calculate the connectivity matrix between these clusters. Is that possible in BrainStorm?

I replied to your other post:

6

Thank you for your response.

I forgot to precise, but I work on intracraniel EEG data. That's why I a able to attribute one electrode to one brain region.

7

I forgot to precise, but I work on intracraniel EEG data. That's why I a able to attribute one electrode to one brain region.

It makes more sense, indeed.
However, in SEEG/ECOG, it might not be a good option to average the signals of different electrodes. Because each channel can contain different information, averaging them might lead to blur the meaningful information.
It could make sense to compute the measure of interest (PSD, connectivity, ...) for each channel individually, and then try to group the final measures by brain region.

8

Thank you a lot.
But maybe one more question then, how can I group the final connectivity measures by brain region using BrainStorm?

9

There are no easy solution for that, you would need to write a Matlab script...