[Eeglablist] Tips to run dipole fit with individualized electrode positions in a batch

[Makoto Miyakoshi]

Dear list subscribers,

I just wanted to share my experience in case you are interested in.

1. Channels names without 10-20 nomenclature (i.e. Fz, Cz, Pz etc) can not
be recognized by 'warp' function in the head model settings. -> Use
fiducials only that have labels such as Nz, LPA, RPA etc that is stored
under EEG.chaninfo.nodatachans

2. Polhemus measurements uses meters whereas EEGLAB default assumes
millimeters. -> x1000 required.

3. If applying individually measured channel locations (using Polhemus,
Zebras, etc) transform parameters should be computed. -> use coregister()
function.

Below is an example.

 % redo channels (transform meter to millimeter!)
    tmpNoDataChans = EEG.chaninfo.nodatchans;
    tmpNoDataChans = rmfield(tmpNoDataChans,'datachan');
    EEG.chanlocs   = [EEG.chanlocs tmpNoDataChans];
    for n = 1:length(EEG.chanlocs)
        EEG.chanlocs(1,n).X = EEG.chanlocs(1,n).X*1000;
        EEG.chanlocs(1,n).Y = EEG.chanlocs(1,n).Y*1000;
        EEG.chanlocs(1,n).Z = EEG.chanlocs(1,n).Z*1000;
    end
    EEG.chaninfo.nodatchans = [];
    EEG = pop_chanedit(EEG, 'eval','chans = pop_chancenter( chans,
[],[]);');

    % compute transform parameter
    [~,transform] = coregister(EEG.chaninfo.nodatchans,
'/data/common/matlab/eeglab/plugins/dipfit2.3/standard_BEM/elec/standard_1005.elc',
'warp', 'auto', 'manual', 'off')

    % manual fitting results
    % [0            -25        -10     -0.1         0     -1.5708     1.15
     1.15      1.1]
    %
    % automatic warping results
    % [-3.4586  -29.9836   -3.8724   -0.0702   -0.0014   -1.5717    1.2789
   1.2018    1.2028]

-- 
Makoto Miyakoshi
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/20140312/d44eec64/attachment.html>