Channel Error

[jokn1902]

Hello,
first, thank you very much for providing your code here. I would like to use it to calculate the Brain Symmetry Index for a chronic stroke subject. I recorded a 64 channel EEG but had to remove some channels during preprocessing so effectively there were 59 channels left the dataset. When I tried to run the code (featx=connectivity_features(x, Fs), all default settings) it produced error: at least two channels needed for analysis. I preprocessed the data in EEGLAB, could this be a problem with the data format? Is there a way to solve this problem? I would be grateful for any suggestions.

Thank you very much, Franziska

[otoolej]

Hi @jokn1902, probably better to generate the BSI through the generate_all_features.m function, by first setting the parameters in the neural_parameters.m file. But can of course do it directly from the connectivity_features.m function too. To do this,

x = randn(2, 10000);
Fs=1;
ch_labels = {'F3', 'F4'};

featx = connectivity_features(x, Fs, 'connectivity_BSI', [], ch_labels);

which will generate the BSI for 2-channels of GWN, 10000 samples long. Is important that x is size (N_channels x N). Also important to specify the channel labels (e.g. in ch_labels), as BSI is calculated from channel pairs on left/right hemispheres.
featx here is 1 x 4 vector, for each of the frequency bands (can set in neural_parameters.m). Make sense?