Artefact Correction

Hi Team,
I came across a recording where the time traces of frontal channels and occipital channels are as follows :

I have observed in previous subjects that, usually such slow periodic components in frontal channels are due to some dental or mascara artefact.
Upon ICA decomposition,

IC1 seems to remove those the periodic component

However, I am not able to figure out, which artefact must have caused such behaviour? As seen, in IC1 topographic plot, there are sources in frontal and occipital. Can you help me understand what could be the possible cause ?

The slow fluctuations might be due to a piece of garment that moves with breathing . Could even be a tattoo that got magnetized if the person got an MRI even weeks before MEG. (The faster ones seem to be captured by IC2 and IC3).

I see.. I am actually removing IC components reflective of eyeblinks and heartbeats for my analysis. Would you advise me to remove the IC 1 or 2,3 in the above discussed subject as part of artefact cleaning process?

It depends on what you would qualify as noise in the circumstances of your study. From their respectives traces, I think IC1 and IC3-4 show non electrophysiological components. IC1 could be replaced by a high pass filter maybe, but again it depends on whether low frequencies are pertinent to your scientific question.

Yes - I will think over these points. For IC3,4: did you infer them as non-electrophysiological components based on the high frequency part (highlighted in yellow) ?

It's possible they're not indeed, given their unusually large amplitude in the original recording.

I see.. Thank you for your guidance !