Dear Vladimir,
Unfortunately my cap is not a standard one that is why I have to use my own coordinates. I changed the polar coordinates to Cartesian coordinates using a radius of 7 cm because EASYCAP company told me it will give me an average head. I used the exact positions of two of my electrodes (54 and 46) as lpa and rpa and used electrode 32 to estimate a position near the nose bridge for my nsa. The first figure in the attached file is the result of my co-registration. And when I use the option of "use headshape points", I get the second figure. Should I change lpa and rpa so that they fall on the red points?
Also, there is one strange thing. When I use the exact xyz coordinates (and I can see that z is positive for the electrodes which have higher positions), it seems that it makes z negative. So, the electrodes fall below the head instead of above it. The third figure shows this situation.
I would appreciate any advice here.
Many thanks,
Pegah
-----Original Message-----
From: Vladimir Litvak [mailto:[log in to unmask]]
Sent: 09 February 2012 19:03
To: Tayaranian Hosseini P.
Cc: [log in to unmask]
Subject: Re: [SPM] 3D electrode positions
Dear Pegah,
If you used standard electrode locations available in SPM (like the
10-20 system) perhaps it'll be better to use the SPM defaults rather than your locations on a sphere as transforming them will introduce more error. If this is some kind of very special cap you can try getting it to XYZ to more or less reasonable size and then use some electrodes that you can locate on the template head as fiducials and do coregistration. SPM will fit the electrodes to the head and distort the sphere if necessary. Spherical head can be used but we don't support it. There is a possibility to use a spherical model but the electrodes need to be in MNI coordinates to use that.
Best,
Vladimir
On Thu, Feb 9, 2012 at 4:01 PM, Tayaranian Hosseini P.
<[log in to unmask]> wrote:
> Dear SPMers,
>
>
>
> I have some 68-channel EEG data (66 channels of EEG and 2 channels of
> EOG), in my recordings, there are 2 more channels for reference (nose
> tip) and ground electrodes. I have the polar locations of the
> electrodes (phi and
> theta) and I can find the xyz locations given a specific head radius
> (assuming head to be sphere).
>
>
>
> 1. According to a standard average head size, what head radius
> do you suggest I use for all my subjects?
>
> 2. Can a spherical head with a fixed radius be a good model for
> source reconstruction and DCM?
>
> 3. What should I do about fiducials when I do not have exact
> measurements? Can I for example use the positions of two of the
> electrodes which are closest to pre-auricular points as lpa and rpa
> and the one closest to nose bridge as nsa?
>
>
>
> Many thanks,
>
> Pegah
|