PLV Connectivity on source

1

Dear sir,

I am working on resting EEG data and I computed the sources by using Minimum source imaging (current density map).

I would like to know for computing PLV connectivity, which scout functions (mean or PCA) should use?

In addition what is the difference between apply scout function Before and After?

I want to use Desikan Killiany atlas.

In advance thank you for your kind attention.
Best regards

1 Like
2

We don't have clear recommendations to give at the moment.
Please browse the forum for existing discussions on the same topic.
Example:

@hossein27en @Sylvain FYI...

3

thank you dear Francois

I have another question : Minimum source imaging ( with current density map). is wMNE( weighted minimum - norm estimation? Or I should use another settings in the source computing to use wMNE for source localization ?

Best,

4

w in wMNE stands for "weighted", if this is your question, to reference the depth-weighting.
The default options of the "Compute sources [2018]" would give you a wMNE solution.
https://neuroimage.usc.edu/brainstorm/Tutorials/SourceEstimation#Advanced_options:_Minimum_norm

1 Like
5

Thank you Francois,

Then without any changes of parameters , I would have wMNE by "Compute sources [2018]".

Is there any manual to know how exactly MNE works ? Because the link in the tutorial doesn't work.

Moreover I have found an article that Compares EEG source reconstructed functions by using brainstorm.

It Probably helps others.
"Comparison of EEG source reconstructed functional networks in healthy subjects elicited during visual oddball task"

6

I fixed the link to the MNE manual.
For further reading, refer to the last section of the tutorial:
https://neuroimage.usc.edu/brainstorm/Tutorials/SourceEstimation#Additional_documentation

7

Hi

I had question about how connectivity measures are calculated at the source level. There are a lot of discussions already on this topic and I thought it better to add to an existing one.

Specifically, I was wondering how these measures (PLV, PTE, etc) are calculated when the option to apply the scout function "After" is selected using a 1xN analysis. My understanding of the pipeline is as follows:

  1. The Hilbert (or other) time-frequency transform is applied to data in each vertex within both scouts
  2. A mean connectivity value for each vertex in scout A is determined (the average connectivity between a vertex in scout A and each vertex in scout B)
  3. The mean (or PCA etc) of these connectivity value within scout A is reported

I assume this is different from the "Before" selection which computes the mean (or PCA etc) activity across the scout first and then applies the transformation and connectivity analysis on this single signal. Is this the correct description of these processes?

Thanks in advance.

8

Specifically, I was wondering how these measures (PLV, PTE, etc) are calculated when the option to apply the scout function "After" is selected using a 1xN analysis

The option before/after defines when the scout function is applied to group multiple source values into one scout value. For the connectivity measures, the logic is the same as for the time-frequency analysis:
https://neuroimage.usc.edu/brainstorm/Tutorials/TimeFrequency#Scouts

The option "after" groups Na x Nb connectivity values for the NxN case (scout A with Na sources, scout B with Nb sources), or Na x 1 values for the Nx1 case,
There is no consensus yet on what the best approach is, therefore we can give recommendations on which option to use.

@Raymundo.Cassani Please make sure this option before/after is correctly documented in the future connectivity tutorial - Thanks!
https://neuroimage.usc.edu/brainstorm/Tutorials/Connectivity2

1 Like