Problem regarding the EEG signals captured
hansamali_paul
Sril Lnkan
in Cyton
I have captured EEG signals with EEG headband kit and Cyton board. When I plot the raw signal with mne I can see an uptrend
in the signal so I tried with the bandpass filter and then I plot the signal with Scipy (Each channel separately). Then I plot the psd plot too. Below are the images of my results. Can I know whether the signal is captured correctly and can I continue with capturing EEG signals?
Comments
The recording files are RAW, NO FILTERS. You must manually apply filters first, such as a bandpass from say .5 Hz to 40 Hz. A notch at your mains frequency is also recommended.
Normal range of EEG (after filtering) is from about 1 microvolt to say 100 microvolts. If you are seeing data stream samples outside that range, then something is wrong. Your Scypy plots should look similar to what you saw in the GUI time series window. The channel plots you post above have huge spikes of .5V or more, (sometimes more than 1V). This may indicate issues with your filtering or plotting.
Below attached file is the raw signal acquired from the openbci device in openbci txt format. no any filtered are applied to the signal. i have attached the screen recording of the GUI also when signal was taken. Can you please tell me whether there is a problem with the signal.
There is NO NEED to submit the same comment or new thread FOUR TIMES. I will delete duplicates.
As stated earlier, EEG is usually below 100 microvolts. In your GUI screenshot, this is the case in the 2nd and 3rd seconds of the trace. However in the 1st and last seconds you have huge amounts of noise. Have you tried applying some common sense detective work to see what is the source of the noise? In other words WHAT is different in those first and last seconds? Is the subject moving or are the cables moving? EEG requires the subject to be fairly still. Is your subject periodically in the vicinity of EMF electromagnetic frequency generating equipment, such as other lab equipment, power supplies, extension cords, lamps, wireless gear, conduits in walls floors ceilings, etc.
Your FFT shows huge noise from your mains frequency, 50 Hz. You MUST turn on the notch filter using the menus.
William
In retrospect, it does look like you have the 50Hz notch. But the mains EMF interference is so huge and wideband, that it is overwhelming the notch filter. The only solution to EMF problems is to get your EEG equipment away from the source of the EMF. There are EMF meters available, and some of these run as apps on phones, provided your phone has the magnetometer peripheral used by 'compass' apps.
https://www.google.com/search?q=emf+meter+phone+apps
https://www.amazon.com/s?k=emf+meters