Hi Vladimir,
For 32 channels that have been highpass filtered, lowpass filtered, epoched and artefact rejected with the following parameters
S = [];
S.D = fullfile(pth, D.fname);
S.badchanthresh = 0.2;
S.methods.channels = {'all'};
S.methods.fun = 'threshchan';
S.methods.settings.threshold = 80; % threshold in microvolts
D = spm_eeg_artefact(S);
the data still contain trials exceeding the threshold (shown for one channel with 1497 epochs). I get the same results when using the GUI.
Best,
Martin
[cid:6159DA1C-1E40-45B7-BD9C-9C055E2750FE@akh06.auh.dk]
On Nov 12, 2010, at 3:35 PM, Vladimir Litvak wrote:
In principle it should work. Could it be that this channel was not
among the ones that were examined or that it was marked as bad? You
can put a breakpoint in spm_eeg_artefact_threshchan line 65 and see
what's happening there. If you can't figure it out, send me an
example.
Vladimir
On Fri, Nov 12, 2010 at 2:25 PM, Martin Jensen Dietz <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Dear Vladimir & meeg'ers,
When using microvolt thresholding for artefact rejection I would assume trials exceeding the specified threshold to be rejected. In this case, that does not seem to be the case for all trials (threshold was set to 70 microvolts). Am I missing something ?
|