<div dir="ltr">Roberto,<div><br></div><div>Aqui algo sobre el cuentionamiento del numero de componentes para realizar ICA, sin embargo en nuestro caso, creo que no sera un parametro sensible ya que tenemos un numero bajo (21) de mediciones</div>
<div><br></div><div>Saludos</div><div><br></div><div>Mauricio</div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Apr 8, 2014 at 8:29 AM, Lozano Soldevilla, D. (Diego) <span dir="ltr"><<a href="mailto:d.lozanosoldevilla@fcdonders.ru.nl" target="_blank">d.lozanosoldevilla@fcdonders.ru.nl</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Fred,<br>
<br>
I don't know the magical number but I see the following options:<br>
<br>
a) Before ICA, concatenate all single trials and ask for the rank of your data. Use it as your cfg.runica.pca input.<br>
<br>
b) Some users notice that when you sort IC by variance, beyond component 30, the IC topo/activation does not look like physiologically meaningfull. It matches with the PCA reduction before ICA with other papers:<br>
<br>
<a href="http://www.ncbi.nlm.nih.gov/pubmed/15219593" target="_blank">http://www.ncbi.nlm.nih.gov/pubmed/15219593</a><br>
<a href="http://www.ncbi.nlm.nih.gov/pubmed/19699307" target="_blank">http://www.ncbi.nlm.nih.gov/pubmed/19699307</a><br>
<br>
Might be somebody in the forum have tried (by simulations or in real data) on the effects of PCA component reduction on ICA.<br>
<br>
I hope it helps,<br>
<br>
Diego<br>
<br>
----- Original Message -----<br>
> From: "Frédéric Roux" <<a href="mailto:f.roux@bcbl.eu">f.roux@bcbl.eu</a>><br>
> To: "FieldTrip discussion list" <<a href="mailto:fieldtrip@science.ru.nl">fieldtrip@science.ru.nl</a>><br>
> Sent: Tuesday, 8 April, 2014 2:46:22 PM<br>
> Subject: [FieldTrip] using cfg.runica.pca to reduce number of ICs<br>
> Dear all,<br>
><br>
> I have a general question relating to the usage of cfg.runica.pca<br>
> during the call to ft_componentanalysis.<br>
><br>
> Is it correct to assume that cfg.runica.pca = length(meg_data.label)<br>
> will<br>
> force the algorithm to return n = length(meg_data.label) ICs, and that<br>
> as a<br>
> result artifacts can be "spread" across several ICs?<br>
><br>
> If that's true, then I imagine that cfg.runica.pac = n/4 will return<br>
> less components<br>
> and reduce the "spread" of artifacts over several components.<br>
><br>
> My question is how to choose the number of principal components to<br>
> which the data<br>
> is reduced before ICA?<br>
><br>
> Best,<br>
> Fred<br>
> ---------------------------------------------------------------------------<br>
><br>
><br>
> _______________________________________________<br>
> fieldtrip mailing list<br>
> <a href="mailto:fieldtrip@donders.ru.nl">fieldtrip@donders.ru.nl</a><br>
> <a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br>
<br>
--<br>
PhD Student<br>
Neuronal Oscillations Group<br>
Donders Institute for Brain, Cognition and Behaviour<br>
Centre for Cognitive Neuroimaging<br>
Radboud University Nijmegen<br>
NL-6525 EN Nijmegen<br>
The Netherlands<br>
<a href="http://www.ru.nl/people/donders/lozano-soldevilla-d/" target="_blank">http://www.ru.nl/people/donders/lozano-soldevilla-d/</a><br>
<br>
_______________________________________________<br>
fieldtrip mailing list<br>
<a href="mailto:fieldtrip@donders.ru.nl">fieldtrip@donders.ru.nl</a><br>
<a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a></blockquote></div><br><br clear="all"><div><br></div>-- <br><div dir="ltr">Javier M. Antelis<br>
</div>
</div>