Error using Uniform Epoch time

Bugs
1

Hi again!
I computed psd sources from a resting EEG file as I used to do it:

Default anatomy
Import the EEG file
set the electrode position
clean the eeg
split the file in epochs
stardardize --> uniform epoch time
No-noise covariance
compute head model (openMeeg)
Sources 2018 (sLoreta)
PSD in 4 bands

everything seems to work, but when I try to display the result I got this message:

Index exceeds matrix dimensions.

Error in bst_memory>GetTimefreqValues (line 2424)
Values = GlobalData.DataSet(iDS).Timefreq(iTimefreq).TF(iRow, iTime, iFreqs);

Error in bst_memory>GetTimefreqMaximum (line 2584)
values = GetTimefreqValues(iDS, iTimefreq, [], [], [], Function);

Error in bst_memory (line 72)
eval(macro_method);

Error in panel_surface>UpdateSurfaceData (line 1698)
TessInfo(iTess).DataMinMax = bst_memory('GetTimefreqMaximum', iDS, iTimefreq, TfInfo.Function);

Error in panel_surface>SetSurfaceData (line 1515)
[isOk, TessInfo] = UpdateSurfaceData(hFig, iTess);

Error in panel_surface (line 40)
eval(macro_method);

Error in view_surface_data (line 298)
isOk = panel_surface('SetSurfaceData', hFig, iTess, OverlayType, OverlayFile, isStat);

Error in tree_callbacks>@(h,ev)view_surface_data([],filenameRelative) (line 2216)
gui_component('MenuItem', jPopup, [], 'Display on cortex', IconLoader.ICON_CORTEX, [],
@(h,ev)view_surface_data([], filenameRelative));

This also happens with 2D disc maps,

everytime I try to compute Welch psd I got messages like these:

BST> process_timefreq: Using 1 windows of 500 samples each
BST> Skipping window #1 because it contains a bad segment.
BST> Skipping window #1 because it contains a bad segment.
BST> Skipping window #1 because it contains a bad segment.
BST> Skipping window #1 because it contains a bad segment.
BST> Skipping window #1 because it contains a bad segment.
BST> process_timefreq: Using 0 windows of 500 samples each
Elapsed time is 0.000708 seconds.

Thanks in advance!

2

BST> Skipping window #1 because it contains a bad segment.

Remove the BAD events from your files.

I computed psd sources from a resting EEG file as I used to do it:

See our recommended pipeline for computing source-level PSD from resting state recordings:
https://neuroimage.usc.edu/brainstorm/Tutorials/RestingOmega

3

The point is: the EEG had no bad segments,
to make sure about this I even made right click and select "accept trials"

4

The trials are not marked as bad trials, but they still include events including the tag "BAD", and are therefore excluded from the PSD computation.

I modified the function executed when clicking on "Accept trial" to remove all the bad events:
https://github.com/brainstorm-tools/brainstorm3/commit/4756493756a7119b0fa02329d1f90a69a88f9643

You can either download the update and do again "Accept trials", or open the trials and delete the BAD events.

5

I modified this behavior, after a remark in this thread:

@tourette95
Can you please check that the proposed modification works for you?
Thanks

6

ok , but I think It is better now