[Eeglablist] signal line noise - an open discussion and debate

[Deepa Gupta]

Greetings community members/signal analyst enthusiasts/clinicians,

I'd like to please open a debate by asking a fundamental yet important
question re. line noise removal step in signal analysis.

   1. Are maximum people using the notch filter or/and 50Hz lowpass filter
   for their signal analysis thereby stripping off other relevant information
   from the signal that may serve the problem statement what-so-it may be by
   following the standard guideline that they/you all may be seeing online or
   in various documentations? Why or why not? Has this been further explored
   by your lab? Please share.
   2. Has anyone ever considered not to remove the line noise entirely by
   chance and just go ahead with proportions or/and taking ratios of the
   signal for obtaining the relevant metrics such as but not limited to the
   qEEG etc.

I ask try to check on this because yes line noise definitely distorts the
signal but even with or without it, the peaks in proportion may serve the
relevant answer for clinical application. Please share with logic as to why
or why not?

Extra share (kind credit to Dr. Makoto and Dr. Scott): Some already
existing tools are but not limited to CleanLine, Spectral interpolation,
BLASST, Zeppelin/plus and more. Which one is it that you use/rely on the
most and why? If you didn't remove line noise at all, that's also fine to
share. Please feel free to comment and elaborate your thoughts if okay.

Greatly appreciate your sharing if possible
Thanking you very much in advance
Sincerely
Deepa



Deepa Gupta <https://urldefense.com/v3/__https://www.linkedin.com/in/deepaguptaresearcher/__;!!Mih3wA!HTYU8bj4mANQ42kd9iJ56nWZdsKI0j3JDho1j9o2ghPsLy8VC8GJipjjH2Vfmmq9trKEZtssMerZtuGzJvbDlzM$ >
Research fellow/Postdoc Scholar, AI-READI/INC, UCSD
Self-funded voluntary founder, https://urldefense.com/v3/__http://www.99p2s.com__;!!Mih3wA!HTYU8bj4mANQ42kd9iJ56nWZdsKI0j3JDho1j9o2ghPsLy8VC8GJipjjH2Vfmmq9trKEZtssMerZtuGzumit0TY$