Should we add more channels for bruxism detection ?

[RobinGuillard]

In fact, we currently use "1" and "2" channels, but we have "1-2", "1-F", "2-F" channels. They are possibly slighly different and may convey more information. As our framework can include as many channels as necessary, should we use these additional channels?

[lkorczowski]

Because of the actual strategy of AmplitudeThresholding (logical AND), I advice to only takes valid channels. When I visualized the signals, I only saw "1" and "2" being valid EMG channels, was I wrong ?

Also:

• Adding channels with artifacts will badly influence output (loosing of a lot of epochs).
• adding redundant channels won't do anything expect making more compational intensive
• I believe "1-F" and other similar channels are bipolar projection of the mentioned channels that convey no real additional information (re-referencing).

[RobinGuillard]

True but channel "1" and "2" are not taking any input from channel "F". What do you think?

[lkorczowski]

What is "F" ?
It doesn't seems to carry any signal.

[lkorczowski]

for subject 1RA17_nuit_hab, try:

• adding electrode 'F' for bruxism detection
• trying to detect on '1' and '2' (try without impedance check).