Question about brain waves output
clatoxen
Italy
Hi, I’m trying to develop a stress index by relating the components of the waves but unfortunately something is wrong. The index assumes a number of values that are out of whack, numerically very small but sometimes with huge spikes and I’m trying to figure out where the problem comes from. In the annex there is the yield of the waves and inanzi all the values what are they? Microvolts? They have a sensible trend or they are totally busted? At what level should I act in the Openbci pipeline to eventually calibrate these values? (I’m not a programmer, I’m learning to use python but I’m at the beginning).
Making comparisons with commercial EEG, under the same conditions, the index of the commercial device is on a scale of 0-100, while with OpenBci the values remain below 40 with sometimes spike that exceed 100 without apparent logic. The formula used is that of the engagment index that relates alpha, beta and theta. I'm using Cyton with only 2 channels recording plus reference and bias.
Thank you for help!
Comments
How did you produce table shown?
https://openbci.com/forum/index.php?p=/discussion/201/large-millivolt-data-values-fbeeg-full-band-eeg
"DC offset" is EEG below .5 Hz. This is removed typically by a bandpass from .5 Hz to say 45 Hz.
We used Processing to add a stress-related formula and generation of the index and wave components in . csv who created this table. So here there are no problems with raw data per se but should I essentially remove the DC component by applying a filter? Where in the pipeline should I apply filtering to have a standard trend? Thanks for the help
Removing the DC offset and applying a notch filter (mains noise) should be your FIRST STEP.