[Mne_analysis] Resting state connectivity

[Peled, Noam]

Hey Denis,
I want to calculate resting state connectivity across time. I'm trying to replicate Linda Douw<https://scholar.google.com/citations?hl=en&user=1YO2IvcAAAAJ&view_op=list_works&sortby=pubdate>'s work, where she correlated between cognitive flexibility and the coefficient of variation (Cv) of the functional connectivity. To calculate the Cv, I need first to calculate the functional connectivity using sliding windows, and assign a value for each window.

Thanks,
Noam

________________________________
From: mne_analysis-bounces at nmr.mgh.harvard.edu [mne_analysis-bounces at nmr.mgh.harvard.edu] on behalf of Denis-Alexander Engemann [denis.engemann at gmail.com]
Sent: Friday, February 03, 2017 4:02 PM
To: sheraz at nmr.mgh.harvard.edu; Discussion and support forum for the users of MNE Software
Subject: Re: [Mne_analysis] Resting state connectivity

Noam,

can you tell me more about your use case? Why would you want to compute connectivity for one window only? If you have some real-time application you would need to create "mini-epochs" of a few samples and then aggregate over them using the epochs-oriented API, but you would run into estimator problems as all these measures scale get more stable with more trials.
With the existing API, single-trial connectivity does not seem easy to realise.

Sheraz,
I have a power envelopes implementation on one of my MNE-Python branches and plan to commit it soon after some final validation is done. Pinging me on this may speed up things. But also here you would go across trials.

Avniel, yes that's a nice paper, they managed to put together some nice materials and making explicit intuitions that had been circulating for quite some time without tapping into the absolute CFC bashing game; indeed they are rather constructive. I also liked their idea that the notion of spurious CFC maybe spurious itself. If CFC is due to harmonics it would at least suggest non-sinusoidal waveforms. Curious to see where this goes.
Denis

On Fri, Feb 3, 2017 at 9:49 PM Sheraz Khan, PhD <sheraz at nmr.mgh.harvard.edu<mailto:sheraz at nmr.mgh.harvard.edu>> wrote:
Hi Noam,

Did you try envelope correlations or its flavors like orthogonal
correlations (Hipp or Brookes) ?

They are found to be most reliable for resting state connectivity.

http://www.sciencedirect.com/science/article/pii/S1053811916301914

HTH

Sheraz




> Yep, this is my problem :)
> Do you know any other approach for resting state connectivity?
>
> Thanks,
> Noam
> ________________________________
> From: mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu>
> [mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu>] on behalf of Denis-Alexander
> Engemann [denis.engemann at gmail.com<mailto:denis.engemann at gmail.com>]
> Sent: Friday, February 03, 2017 2:22 PM
> To: Discussion and support forum for the users of MNE Software
> Subject: Re: [Mne_analysis] Resting state connectivity
>
> Ah no the assumption here is that you aggregate over trials. If you just
> send one window you cannot get a meaningful score.
>
> On Fri, Feb 3, 2017 at 8:13 PM Peled, Noam
> <NPELED at mgh.harvard.edu<mailto:NPELED at mgh.harvard.edu><mailto:NPELED at mgh.harvard.edu<mailto:NPELED at mgh.harvard.edu>>> wrote:
> Hey Denis!
> Thanks for the answer!
> My problem isn't creating the events (actually I didn't know about this
> function and now I can delete the code I wrote for that), but how to send
> one window to the mne.connectivity.spectral_connectivity (one of the PLI
> methods) without getting all zeros or nans.
>
> Thanks,
> Noam
> ________________________________
> From:
> mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu><mailto:mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu>>
> [mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu><mailto:mne_analysis-bounces at nmr.mgh.harvard.edu<mailto:mne_analysis-bounces at nmr.mgh.harvard.edu>>]
> on behalf of Denis-Alexander Engemann
> [denis.engemann at gmail.com<mailto:denis.engemann at gmail.com><mailto:denis.engemann at gmail.com<mailto:denis.engemann at gmail.com>>]
> Sent: Friday, February 03, 2017 1:51 PM
> To: Discussion and support forum for the users of MNE Software
> ý[mne_analysis at nmr.mgh.harvard.edu<mailto:mne_analysis at nmr.mgh.harvard.edu><mailto:mne_analysis at nmr.mgh.harvard.edu<mailto:mne_analysis at nmr.mgh.harvard.edu>>]ý
> Subject: Re: [Mne_analysis] Resting state connectivity
>
> Hi Noam,
>
> checkout `mne.make_fixed_length_events`,
>
> Denis
>
> On Fri, Feb 3, 2017 at 7:50 PM Peled, Noam
> <NPELED at mgh.harvard.edu<mailto:NPELED at mgh.harvard.edu><mailto:NPELED at mgh.harvard.edu<mailto:NPELED at mgh.harvard.edu>>> wrote:
> Hey all,
> What is the best approach for resting state connectivity analysis?
> It seems that mne.connectivity.spectral_connectivity is design for evoked
> data.
> In resting state, the different 'epochs' are sliding windows in time, so
> for each calculation I can send only one epoch, which yields to zeros
> (pli2_unbiased).
>
> Thanks,
> Noam
>
>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu><mailto:Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu>>
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>
>
> The information in this e-mail is intended only for the person to whom it
> is
> addressed. If you believe this e-mail was sent to you in error and the
> e-mail
> contains patient information, please contact the Partners Compliance
> HelpLine at
> http://www.partners.org/complianceline . If the e-mail was sent to you in
> error
> but does not contain patient information, please contact the sender and
> properly
> dispose of the e-mail.
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu><mailto:Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu>>
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>
>
> The information in this e-mail is intended only for the person to whom it
> is
> addressed. If you believe this e-mail was sent to you in error and the
> e-mail
> contains patient information, please contact the Partners Compliance
> HelpLine at
> http://www.partners.org/complianceline . If the e-mail was sent to you in
> error
> but does not contain patient information, please contact the sender and
> properly
> dispose of the e-mail.
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu>
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis


-------------------------
Sheraz Khan, M.Eng, Ph.D.
Instructor in Neurology

Athinoula A. Martinos Center for Biomedical Imaging
Massachusetts General Hospital
Harvard Medical School

McGovern Institute for Brain Research
Massachusetts Institute of Technology

Tel:   +1 617-643-5634<tel:+1%20617-643-5634>
Fax:   +1 617-948-5966<tel:+1%20617-948-5966>
Email: sheraz at nmr.mgh.harvard.edu<mailto:sheraz at nmr.mgh.harvard.edu>
       sheraz at mit.edu<mailto:sheraz at mit.edu>
Web:   http://sheraz.mit.edu
_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu>
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20170205/196babcb/attachment-0001.html