[Eeglablist] Clustering components of different groups of subjects with several tasks

[Makoto Miyakoshi]

Dear Aleksandra,

That sounds like a limitation of this approach, so no perfect answer.

Alternatively, you can make a relatively big clusters and show dipole
density difference. I'll give you my plugin (which is alpha version now)
for that purpose. Email me if you are interested in.

You may also want to try Measure Projection to see what it returns. It does
not generate too many domains (compared to k-means clustering where we use
10-25) so you'll have no choice there and solve the problem.

Makoto

Sent from my Brain

2013/2/27 Aleksandra Vuckovic <Aleksandra.Vuckovic at glasgow.ac.uk>

> Dear Makoto,
> I do expect different spatial distribution of some dipoles in different
> groups, and some clusters existing dominantly in one group only. If I use
> smaller number of clusters I might miss it. Which clustering method would
> you recommend?
> Many thanks
> Aleksandra
>
> Sent from my iPhone
>
> On 27 Feb 2013, at 19:02, "Makoto Miyakoshi" <mmiyakoshi at ucsd.edu> wrote:
>
> Dear Aleksandra,
>
> > I can hardly get the same number of subjects from each group,
>
> Why don't you use smaller number of clusters? I don't see any necessity to
> create separate STUDY in your case.
>
> Makoto
>
> 2013/2/27 Aleksandra Vuckovic < <Aleksandra.Vuckovic at glasgow.ac.uk>
> Aleksandra.Vuckovic at glasgow.ac.uk>
>
>>   Hi
>> I have three groups (patients) and three conditions, 9 variables in
>> total. I am trying to cluster their ICA components and wonder if it is a
>> good idea to create only one study. Different groups have (as
>> expected) slightly different areas of cortical activation, so when I
>> create a common cluster I can hardly get the same number of subjects from
>> each group, so further comparison between conditions may be affected by
>> different number of subjects in each group.
>>
>> Has anybody faced a similar problem? Should I create separate studies
>> for separate groups and then compare coordinates of similar clusters
>> (does not seem like an optimal solution)
>>
>> Many thanks,
>> Aleksandra
>>
>> _______________________________________________
>> Eeglablist page: <http://sccn.ucsd.edu/eeglab/eeglabmail.html>
>> http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> <eeglablist-unsubscribe at sccn.ucsd.edu>
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> <eeglablist-request at sccn.ucsd.edu>eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> Makoto Miyakoshi
> JSPS Postdoctral Fellow for Research Abroad
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
>


-- 
Makoto Miyakoshi
JSPS Postdoctral Fellow for Research Abroad
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20130227/d66b43f6/attachment.html>