Hello
We start a project on the cortico-muscular coherence, using EEG and EMG signals. However, we don't get coherence between 0 and 1 (see Image). Also, we apply it to an EMG electrode when we would like to apply it to a differential EMG montage (Elec1 - Elec 2). Thanks for your help
EEG_PSD_Subject01_Vincent_marche_2min mscohere_1.0Hz_45win__EXG1__EEG_sensors.tif (169.7 KB)
However, we don't get coherence between 0 and 1 (see Image).
This is an issue with the display: the units are shown as being milliVolts instead of Coherence.
The values along the axes are to be interpreted as /1000, so they are all between 0 and 1 on this graph.
@Raymundo.Cassani Can you please edit bst_connectivity to save correctly the field DisplayUnits in the output file?
we apply it to an EMG electrode when we would like to apply it to a differential EMG montage (Elec1 - Elec 2)
One simple approach could be to use the process "Pre-process > Run Matlab command" to compute this difference and save it in our of the channels.
% With iEMG1 and iEMG2 the indices of the two EMG channels in the channel file.
Data(iEMG1,:) = Data(iEMG1,:) - Data(iEMG2,:);
Hello,
Thank you for your answer. However, we still have another problem. We have replaced the variables in the code by electrodes that are in our data and when we're launching the code it shows us an error (see image).
You need to enter the INDEX of each channel that you are expecting to use, not its name.
To get the list of channels with their indices: right-click on the channel file > Edit channel file.
Issue is solved in: Bugfix cnx display units (#610) · brainstorm-tools/brainstorm3@e4c248f · GitHub 
@dex, update and restart Brainstorm to see the changes
Hello,
I have another problem concerning the cortico-muscular coherence, the values obtained are in microvolt and some times in millivolt, the score which can be considered as on /1000 in millivolt must be also in microvolt on /10000?
(See pic 1)
And the values obtained in microvolt exceed 10,000 microvolts.
(See pic 2)
Thank you for your help.
Hi @dex,
Coherence spectrum was wrongly displayed with the sensor units (in this case mV and uV).
This is now fixed (previous message).
Just update Brainstorm, and open again the coherence result. This is how it should look:
Hello, thank you for your answer.
