[Eeglablist] How can I use 128 Biosemi channel locations in DIPFIT?

[Alejandro]

Hi James,

I have some comments on this issue.

Normally one would warp the individual electrode positions down to the 
standard MNI head model used by Dipfit, this can be done based on some 
landmarks (set of common electrodes or fiducials). The automatic 
  warping of the individual montage to the MNI space provides an 
anatomical location for your sensors in that common space. The warping 
is done according to some mathematical criteria trying to preserve the 
relation between individual sensors so they get affected in the same way 
by the warping procedure.

Now, if the automatic coregistration does not work then you would need 
to do it manually, this pathway could be really painful, and could 
introduce non-optimal locations because the fitting criteria is 
basically your eyes.

Here is my advise, if you find yourself warping a bunch of electrodes by 
hand you may consider using the standard electrode positions that 
correspond to the montage you have, they are already in MNI's space so 
no warping is needed. This aren't the true locations but will be 
probably better than the ones you have warped by hand, at the end of the 
day, the MNI head model is just a defined template that represents the 
head of all your subject, but could be very far from any real head.

In the folder above you can find different electrode position files for 
different standard montages, 10-20 among them:
/.../eeglab/plugins/dipfit2.2/standard_BEM/elec

There is a third solution though, this one uses the cortical surface of 
a template (also in MNI space) to warp the head model of the template to 
the individual channel positions ( you have to provide at least 3 
fiducials, that's it), resulting in an "individualized" head model, now, 
the results of this program cannot be used by Dipfit without making tons 
of modifications in Dipfit's code, however, it can be used by other 
inverse solution methods we're developing at SCCN (so far is looking 
good). This is not a plugin yet and still in active development, but 
you're welcome to try the code if you want to.

Regards,
Alejandro




From:*Makoto Miyakoshi*<mmiyakoshi at ucsd.edu <mailto:mmiyakoshi at ucsd.edu>>
Date: Wed, Aug 15, 2012 at 11:29 AM
Subject: Re: [Eeglablist] How can I use 128 Biosemi channel locations in 
DIPFIT?
To: James Schaeffer <schaefj3 at gmail.com <mailto:schaefj3 at gmail.com>>
Cc:eeglablist at sccn.ucsd.edu <mailto:eeglablist at sccn.ucsd.edu>


Dear James,

This is from my experience of collaboration. I was using MNI model.

1. My conclusion is that you must have 10-5 labels to warp electrodes.
I remember I tried to pair the channels that did not have 10-5 labels
but did not work.

2. So what I did was to pick up 5-10 electrodes that had closest
positions to 10-5 and rename their channels with (approximately)
corresponding 10-5 label names. Using those channels, I could warp
electrodes to fit the model.

3. Of course, you can warp your channels manually. I'm not sure about
availability of fiducials now (I knew there were some issues in it,
but it could have been solved by update. If you find problems please
let us know). The only drawback of this manual adjusting is that it
takes time. However, if your channel locations are not measured but
just using MNI 10-5 template AND your head model is standard MNI
template and not subject' MRI, then you can use parameters obtained
from one subjects for all other subjects.

I know this head model setting has many issues, and sometimes
solutions are not pretty straightforward. Sorry about that. We'll
support you so just let us know when you encounter problems.

Makoto


2012/8/13 James Schaeffer <schaefj3 at gmail.com <mailto:schaefj3 at gmail.com>>:
>  Dear eeglablist
>
>  I am trying to use 128 Biosemi channel locations in DIPFIT, but I am 
having
>  trouble aligning my electrodes to match the DIPFIT head model.  Do the
>  biosemi locations correspond with the 10-20 system?  If so, does 
anyone have
>  the 10-20 labels?  If not, how can I work around this?  I have tried
>  aligning the fiducials but the channel locations appear far too low on the
>  head model and need to be manually moved into place.  If I move all the
>  locations up about 60 mm, they look like they are in place.  Has 
anyone else
>  experienced this? Is there a better way to do this?  Any help would be
>  greatly appreciated.
>
>  Thanks in advance,
>  James
>
>  _______________________________________________
>  Eeglablist page:http://sccn.ucsd.edu/eeglab/eeglabmail.html 
<http://sccn.ucsd.edu/eeglab/eeglabmail.html>
>  To unsubscribe, send an empty email 
toeeglablist-unsubscribe at sccn.ucsd.edu 
<mailto: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 <mailto:eeglablist-request at sccn.ucsd.edu>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20120815/cab0b0d9/attachment.html>