DICS

[Muthuraman Muthuraman]

Hello

Thankyou for the suggestions
I have dowloaded the latest version now and i am working with it
i have created the own elec file which is "elec1" and these are the steps i
follow
please go through it and reply me, why is the error

-First i calculate the coherence using the freqanalysis and freqdescriptives
which works perfectly
-Second i calculate the Grid using Prepare_leadfield

vol   = [];
vol.r = 12 * [0.88 0.92 1.00];
vol.c = [1 1/80 1];
vol.o = [0 0 0];
>>cfg1            = [];
cfg1.elec       = elec1;
cfg1.vol        = vol;
cfg1.resolution = 2;
grid = prepare_leadfield(cfg1);

which also works i get the output for the Grid

-Third I want use the source analysis for the DICS

cfg.method='dics';
cfg.grid=grid;
cfg.elecfile=elec1;
cfg.vol=vol;
[source]=sourceanalysis(cfg,freqmtmfft);
using headmodel specified in the configuration
??? Error using ==> fprintf
Function is not defined for 'struct' inputs.

Error in ==> fieldtrip-20060712\private\prepare_vol_sens at 310
  fprintf('reading electrodes from file %s\n', cfg.elecfile);

Error in ==> sourceanalysis at 574
[vol, sens, cfg] = prepare_vol_sens(cfg, data);




-I also tried grid using the freqmtmfft-which as the results for the
freqanalysis
vol   = [];
vol.r = 12 * [0.88 0.92 1.00];
vol.c = [1 1/80 1];
vol.o = [0 0 0];
>>cfg1            = [];
cfg1.elec       = elec1;
cfg1.vol        = vol;
cfg1.resolution = 2;
grid = prepare_leadfield(cfg1,freqmtmfft);

I get error

using headmodel specified in the configuration
using electrodes specified in the configuration
??? Error using ==> delaunayn
Not enough unique points to do tessellation.

Error in ==> delaunay at 49
    tri = delaunayn([x(:) y(:)]);

Error in ==> fieldtrip-20060712\private\prepare_vol_sens at 330
  elec.tri   = delaunay(elec.prj(:,1), elec.prj(:,2));

Error in ==> prepare_leadfield at 179
[vol, sens, cfg] = prepare_vol_sens(cfg, data);


Thanking you

with regards
Muthuraman.




>From: Robert Oostenveld <r.oostenveld at FCDONDERS.RU.NL>
>Reply-To: FieldTrip discussion list <FIELDTRIP at NIC.SURFNET.NL>
>To: FIELDTRIP at NIC.SURFNET.NL
>Subject: Re: [FIELDTRIP] DICS
>Date: Tue, 11 Jul 2006 17:51:28 +0200
>
>Dear Muthuraman
>
>On 11 Jul 2006, at 16:59, Muthuraman Muthuraman wrote:
>>To use the DICS method
>>
>>We have a 64 channel EEG measurement from (Neuroscan) and 4 EMG  channels
>>recording
>>and also we have the 64 channel measurement in FRMI
>>
>>the electrode locations we have it in a file with 3 columns x,y,z  for
>>each electrode
>>using the polhemus sensors but our own software
>>
>>In the Source analysis
>>Cfg.method='DICS';
>>but Cfg.xgrid, ygrid and zgrid
>>the grid values are the sensor location values
>>
>>or it should be the
>>cfg.grid.pos-where i can give in the 3 values [x y z] for each  electrode
>
>in SOURCEANALYSIS you specify using cfg.xgrid/ygrid/zgrid the  locations at
>which the _dipoles_ will be placed, it is not the  position of the
>electrodes. At each grid location a spatial filter is  constructed and the
>power (and coherence) is computed. If you do not  want to scan on a full
>3-D grid, but on only a few dipole locations  of interest, you can use
>cfg.grid.pos to specify those dipole locations.
>
>The electrode positions are specified in the cfg.elec structure, or  in the
>cfg.elecfile filename. You can find more details on the  definition of
>electrode positions at http://www2.ru.nl/fcdonders/ fieldtrip/doku.php?
>id=fieldtrip:documentation:frequently_asked_questions#how_are_electrodes
>_magnetometers_or_gradiometers_described
>
>Furthermore, you can try using the READ_FCDC_ELEC function for  reading the
>electrode locations, but probably taht will not work  since youy are using
>your own software for digitization.  Alternatively, you can manually
>construct an "elec" electrode  definition.
>
>best regards
>Robert
>
>PS please upgrade to a recent fieldtrip version, the sourceanalysis
>options cfg.xgrid/ygrid/zgrid have been renamed. The old options  still
>will work, but in general it is preferable to work with a  recent copy of
>the toolbox.