Hello!
I am working with 32-channel EEG data with 1 additional ECG channel (Fs = 1000 Hz for both signal types). While I am able to see cardiac artifacts in plots from ECG epochs/evokeds objects, I don’t see any independent component that is significantly correlated with the ECG signal; the scores I get from ica.find_bads_ecg are very low (e.g., <0.25).
Therefore, I am considering using SSP for cardiac artifacts removal. My concern/question is that whether there are any reservations against using SSP for cardiac artifacts when I am using ICA for blink artifacts? If not, what is the right order of ICA/SSP application if it matters?
Thanks,
Parth