[Eeglablist] Question about importing brain vision recorder files, reference issues

[Thomas, Robin]

Hello all,

My collaborator has an actichamp 32 channel system that records EEG using
Brain Vision Recorder.  For … reasons …, the reference channel during
recording is set to CP2 (# 23 on their electrode map).  When I import the
data into EEGLAB, the data set has 31 channels in it (which does not
include the CP2 channel that was used as a reference).  I want to add back
the reference channel as CP2 with all 0’s in it so that I can re-reference
to the average and get data back into that location offline.  There doesn’t
seem to be a way to do that using the gui edit channel locations, or by
using the re-reference option in tools.  I have knowledge of the channel
locations in spherical coordinates (theta, phi, radius) that I could
manually input as well (the locations file from the manufacturer has to
many channels in it - it is an xml file - and if I try to read it in using
the edit locations gui, errors abound).   I am willing to edit variables
directly using the command line to achieve what I need.


For example, suppose there are 503990 time points (i.e., columns) in the
EEG.data matrix when there are only 31 channels imported.  I am thinking
that I could do this:


EEG.data(32,:) = zeros(1,503990);

EEG.nbchan = 32;


Now the EEG structure looks like this:


EEG =


  struct with fields:


             setname: 'TestHPFilt'

            filename: ''

            filepath: ''

             subject: ''

               group: ''

           condition: ''

             session: []

            comments: 'Original file: test9-3-2021.eeg'

              nbchan: 32

              trials: 1

                pnts: 503990

               srate: 500

                xmin: 0

                xmax: 1.0080e+03

               times: [1×503990 double]

                data: [32×503990 single]

              icaact: []

             icawinv: []

           icasphere: []

          icaweights: []

         icachansind: []

            chanlocs: []

          urchanlocs: []

            chaninfo: [1×1 struct]

                 ref: 'common'

               event: [1×243 struct]

             urevent: [1×243 struct]

    eventdescription: {''  ''  ''  ''  ''  ''  ''  ''  ''}

               epoch: []

    epochdescription: {}

              reject: [1×1 struct]

               stats: [1×1 struct]

            specdata: []

          specicaact: []

          splinefile: ''

       icasplinefile: ''

              dipfit: []

             history: '↵EEG.etc.eeglabvers = '2021.1'; % this tracks which
version of EEGLAB is being used, you may ignore it↵EEG = pop_loadbv(‘path
name deleted for privacy’, 'test9-3-2021.vhdr', [1 503990], [1 2 3 4 5 6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31]);↵
EEG.setname='Test';↵EEG = eeg_checkset( EEG );↵EEG = pop_eegfiltnew(EEG,
'locutoff',0.5,'plotfreqz',1);↵EEG.setname='TestHPFilt';↵EEG =
eeg_checkset( EEG );↵pop_eegplot( EEG, 1, 1, 1);↵EEG = eeg_checkset( EEG );↵EEG
= eeg_checkset( EEG );'

               saved: 'no'

                 etc: [1×1 struct]

                 run: []

             datfile: ''




But then,I can’t seem to use edit channel locations anymore, when I select
that function from the tools menu and accept the default (channels have
known labels), I get one channel (E1) and everything seems missing.   Is
there a good way to add the recording reference channel data (all 0’s) into
the data array and re-reference to the average to get data back into CP2
location?  I am not opposed to doing all of this ‘by hand’ as in directly
editing using a script, e.g., these fields in the EEG struct.

thank you in advance,
Robin
-- 
Robin D. Thomas, Ph.D.
Full Professor and Director
Center for Human Psychophysiology
Program in Brain and Cognitive Science
Miami University, Oxford, OH 45056