seeking advice re: EEG raw data validity

hazyliumhazylium Singapore
in General Discussion

My team has been collecting EEG data using the Cyton 8-channel board with IDUN dryodes (we constructed our own headset to hold the EEG dryodes in place on the subject's head). We are new to EEG reading, so we are not sure if the raw data is within acceptable range or if the readings are just noise from the scalp.

Are the readings in this file (OpenBCI-RAW-2021-12-12_22-06-50.txt) already in µV or do we have to perform any calculations or conversion?

For reference: in our region, the power grid frequency is 50 Hz so we have already filtered it out and also bandpass from 0.5 to 100 Hz.

This is our preprocessed data (pre_zscore_preprocessed_data.csv) before zscore normalization using the raw data from the OpenBCI recordings. Are these values within range?

Comments

  • wjcroftwjcroft Mount Shasta, CA

    https://openbci.com/forum/index.php?p=/discussion/201/large-millivolt-data-values-fbeeg-full-band-eeg

  • wjcroftwjcroft Mount Shasta, CA

    @hazylium said:
    ...
    For reference: in our region, the power grid frequency is 50 Hz so we have already filtered it out and also bandpass from 0.5 to 100 Hz.

    With two samples per waveform, 250 hz sample rate would record (very poorly) a 125 hz sine wave. Only two samples per wave. More conservative and safe would use 5 or more samples per wave: 250/5 = 50 hz.

    With only 2 samples per waveform, any analog to digital converter is subject to severe 'aliasing' effects, since shifting the sample points 90 degrees will result in close to zero measured amplitudes. (The zero crossings vs. the peaks.)

    Hence your bandpass could be .5 Hz to 45 Hz, since 50 Hz is already your notch filter.

    Regards, William

  • hazyliumhazylium Singapore

    Hi William,

    Thanks v much for your replies! That was really helpful.

    I have a follow-up question: could you advise me on how to change the physical frequency of the EEG recording?

  • wjcroftwjcroft Mount Shasta, CA

    Hello Hazylium,

    The 'real-time' streaming from the Cyton is limited to 250 Hz because of the radio packet bandwidth. It may be possible to increase the sample rate when recording to SD card. However there have been some past discussions on that subject on the Forum. Use the "Google Advanced Search" button in the upper right column.

    https://docs.openbci.com/Cyton/CytonDataFormat/

    William

Sign In or Register to comment.

© Vanilla Keystone Theme 2026