[FieldTrip] Resting State EEG Connectivity

[Eelke Spaak]

Hi Elene,

I think ft_sourceparcellate should also work on the input timecourses.
(Though I'm not 100% sure about that.) You could also consider using a
lower-resolution grid for your source analysis, and/or combining the time
courses for different positions manually before the call to
ft_connectivityanalysis.

Good luck!
Eelke

On Wed, 8 May 2019 at 11:52, Elene Beitia Loinaz <
elene.beitia at alumni.mondragon.edu> wrote:

> Dear Eelke,
>
> We are following the tutorial that you have send us but we want to perform *ft_sourceparcellate
> *before connectivity to group the centroide of the voxels into regions,
> since we want to reduce the grid. Otherwise, computing the lines requieres
> computationally big effort which causes MEMORY errors.
>
> [image: image.png]
>
> The problem cames because in our *source*, which is obtained from* ft_sourceanalysis
> ,* the number of* source.pos *is really big and matlab does not work with
> this. We have seen that if we increase the size of the grid when we create
> the leadfield we do not have Memory problems, but we think that is not very
> accurate. That is why we want to create before *ft_connectivityanalysis *different
> regions to perform connectivity analysis between them and do not have
> computatioinal probems.
>
> Thanks for your help.
>
>
> Hau idatzi du Eelke Spaak (e.spaak at donders.ru.nl) erabiltzaileak (2019
> mai. 8, az. (11:59)):
>
>> Dear Elene,
>>
>> Have a look at ft_sourceparcelllate and the following tutorial,
>> particularly the part about parcellation:
>> http://www.fieldtriptoolbox.org/tutorial/networkanalysis/ .
>>
>> Cheers,
>> Eelke
>>
>>
>> On Wed, 8 May 2019 at 08:57, Elene Beitia Loinaz
>> <elene.beitia at alumni.mondragon.edu> wrote:
>> >
>> > Dear all,
>> >
>> > Continuing with the connectivity issue, we have seen that the problem
>> of memory come from the size of the leadfield`s grid used previously for
>> source reconstruction and now for connectivity. The variable related with
>> that grid is data.pos,  which was created in ft_prepareleadfield. This
>> value contains the coordinates of the centroids' voxels, related with the
>> grid. As this variable is too large, it supposes really high computational
>> effort that matlab cannot handle.
>> >
>> > We want to cluster the data.pos voxels coordinates into different brain
>> regions such as: temporal area, frontal field....
>> >
>> > Does anyone know how we can group those coordinates into regions in
>> order to decrease the size of data.pos variable needed in
>> ft_connectivityanalysis function?
>> >
>> > Thank you in advanced.
>> >
>> > Hau idatzi du Elene Beitia Loinaz (elene.beitia at alumni.mondragon.edu)
>> erabiltzaileak (2019 mai. 7, ar. (11:11)):
>> >>
>> >> Hi Fieldtripers,
>> >>
>> >> We are trying to perform a conectivity analysis using source
>> reconstructed data. We have seen that ft_connectivityanalysis can be
>> calculated using the output of ft_freqanalysis or ft_sourceanalysis.
>> Someone knows which are the benefits of using the results obtain in
>> ft_sourceanalysis to perform the connectivity analysis?
>> >>
>> >> Furthemore, we have problems when we compute the choerence approach
>> with the source reconstructed data.
>> >>
>> >> Our code:
>> >>
>> >> cfg           = [];
>> >> cfg.method    = 'coh'
>> >> coh           = ft_connectivityanalysis(cfg, beamformer);
>> >>
>> >> The error that we obtain is:
>> >> Out of memory. Type "help memory" for your options.
>> >>
>> >> Error in univariate2bivariate (line 314)
>> >>          data.crsspctrm = (transpose(mom)*conj(mom))./nrpt;
>> >>
>> >> Error in ft_connectivityanalysis (line 449)
>> >> [data, powindx, hasrpt] = univariate2bivariate(data, 'mom',
>> 'crsspctrm', dtype, 'cmb', cfg.refindx,'keeprpt', 0);
>> >>
>> >> This segment of code has been concluded following the tutorials below:
>> >> http://www.fieldtriptoolbox.org/tutorial/networkanalysis/
>> >> http://www.fieldtriptoolbox.org/tutorial/connectivity/
>> >>
>> http://www.fieldtriptoolbox.org/workshop/meg-uk-2015/fieldtrip-connectivity-demo/
>> >>
>> >> Thank you in advanced,
>> >>
>> >> Waiting for your answer.
>> >
>> > _______________________________________________
>> > fieldtrip mailing list
>> > https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>> > https://doi.org/10.1371/journal.pcbi.1002202
>>
>> _______________________________________________
>> fieldtrip mailing list
>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>> https://doi.org/10.1371/journal.pcbi.1002202
>>
> _______________________________________________
> fieldtrip mailing list
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> https://doi.org/10.1371/journal.pcbi.1002202
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20190508/6c8a9deb/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 9266 bytes
Desc: not available
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20190508/6c8a9deb/attachment-0002.png>
-------------- next part --------------
_______________________________________________
fieldtrip mailing list
https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
https://doi.org/10.1371/journal.pcbi.1002202