Hi Francois and Martin,
I recently read a paper about PLV (Bruña et al., 2018) that proposes a new, faster and less biased way of computing PLV, even when assessing whole-brain connectivity. The paper includes their Matlab code.
I know your to-do list is long but could this be included in Brainstorm at some point?
Here is the link to the paper
Many thanks!
Thomas
Dear Thomas,
Many thanks for your message.
As you mentioned, we have other priorities, so we cannot add this functionality to the Brainstorm at this time. Please consider that adding a new function needs several steps like necessity, validating the algorithm and simulations.
As you know, you can write your own process function based the available guidelines
in our tutorial.
https://neuroimage.usc.edu/brainstorm/Tutorials/Scripting
I completely understand. Thanks for the fast answer Hossein!
Please correct me if I'm wrong but for iPLV simply open bst_connectivity.m and change abs to imag:
% Measure
if strcmpi(OPTIONS.Method, 'plv') || strcmpi(OPTIONS.Method, 'plvt')
% Apply measure
switch (OPTIONS.PlvMeasure)
case 'magnitude'
FileMat.TF = abs(FileMat.TF);
FileMat.Measure = 'other';
otherwise
FileMat.Measure = 'none';
end
else
FileMat.Measure = 'other';
end
That sounds promising, I hope your'e right!
Just seeing this now, but since this summer, the PLV implementation follows the Bruña and Pereda's paper, and it's indeed faster 
I added the basic one though, not the ciPLV (which is indeed very easy to modify).
What's the best way of adding the ciPLV? @Francois @Sylvain ?
If we decide that it's the default and that PLV becomes ciPLV, then it's easy for me.
Otherwise if we have to add a new measure, better I send you the modification directly.
cheers
I think it would be essential to have one PLV and one iPLV estimate. ciPLV seems to be the best for the latter.
It would be great to add these connectivity measures to BS!!
Hopefully
@danielemarinazzo
Yes, please send me your code, in a way that is compatible with the current implementation of PLV:
https://github.com/brainstorm-tools/brainstorm3/blob/master/toolbox/connectivity/bst_connectivity.m#L505
Ideally, you should be able to replace the case 'plv' with your code and it should work.
I will add your code to a new case ciplv and add everything else needed in the rest of the software so that the output files are recognized correctly everywhere.
Thanks!
Hi, I have posted the edits in a pull request on github, my previous replies were too noisy and not all consistent with the brainstorm format.
The implementations are in the commented lines in the pull request here
Thank you!
I haven't forgotten about your request. We will discuss this with the rest of the Brainstorm team at our biweekly meeting next week. Then I will need a few extra days to implement the new option in the PLV process.
Please be patient
Thanks, and no hurry at all!
As a matter of fact in this case the delay was beneficial so I could double check the code and correct some inconsistencies.
Please feel free to ping me in case you need something when you discuss the implementation.
I opened a PR to discuss the implementation:
https://github.com/brainstorm-tools/brainstorm3/pull/497
Can you please comment directly in there?
Thanks
Merged here:
Thank you very much!!! 