Cardiac Artifact

knightv8knightv8
edited May 2016 in Hardware
Hi,
I am a PhD student using OpenBCI.

I cannot seem to get the Cardiac artefact (EKG/ECG and Pulse) to show up, on a referential montage (FP1 - A1, FP2 - A1, C3 - A1 etc). This should be the ideal set-up to be able to show this artefact. Is this artefact automatically removed by the 'common mode noise rejection' ?

I do know how to set-up the equipment to get an EKG signal, but I am interested in the artefact that is usually displayed on almost all electrodes in a referential montage.

If the issue is the 'common mode noise rejection' is there a way to bypass this, to show this artefact for classification reasons.

Thanks
Patrick


Comments

  • wjcroftwjcroft Mount Shasta, CA
    Patrick, hi.

    So to clarify, your question is really "Cardiac Artifact expected in referential EEG?"

    Can you give a link or reference mentioning this phenomena? I've not seen it myself on the several types of EEG equipment I've used; including OpenBCI.

    There shouldnt be much EKG electrical activity between points on the head or head / ear. If there is any signal, it would be more canceled by the differential amp between the reference and active; versus canceled by the Bias / Ground.

    William

  • knightv8knightv8
    edited May 2016
    Hi William,
    many thanks for your reply.

    Yes you are right re:question title; I should have explained it better.

    Let me give you a few links:


    Page 49-50


    etc.

    Thanks
    Patrick
  • wjcroftwjcroft Mount Shasta, CA
    Interesting. The LinkedIn article is good. Hmm, according to that first screenshot, the ECG artifacts at Fp1 are about 30 uV in size; have never seen anything that large. On second consideration I admit I have seen very occasional cardiac artifacts in EEG, but it's usually a rare phenomena. Not something you can 'reliably' expect to pickup. The LinkedIn does point out that:

    "Essentially, the artifact is a poorly formed QRS complex.Most prominent when the neck is short.The complex usually is diphasic, but some EEGs may depict it as either monophasic or triphasic."

    William

  • wjcroftwjcroft Mount Shasta, CA
    In addition to minimizing this effect using averaged reference, it's also minimized using "linked ears", which is the default montage for QEEG, and what many neurofeedback practitioners use (including myself).
  • knightv8knightv8
    Hi William,
    thanks once again for your prompt replies.

    Yes I though so as well re:article, but I have read so many books lately, I get mixed up who said what :)

    I think that those where using clinical EEG equipment, which would differ from our equipment. Maybe that is the case for the 30uV and for it being easily detectable?

    Yes I understand regarding reliability to pick it up. However I have tried with several (10) different subjects with various ethnicity and body types but not luck; not even once. Do you recall if you have seen this occasional artefact using OpenBCI or another EEG device?

    Just as a reference (have read it somewhere) that both linked ears must detect the heart beat to cancel each other out and remove the EKG artefact.

    Many thanks
    Patrick
  • wjcroftwjcroft Mount Shasta, CA
    I actually have not seen this with OpenBCI. I did see it once with my Pocket-Neurobics amp, but in that case I was also using a pressure based (velcro strap) system for holding on the electrodes. In that situation where the pulse was showing up, it turned out I had an electrode positioned right over a prominent arterial pulse point. After moving the electrode slightly, the effect disappeared.

    With linked ears you get a more 'averaged' ear reference as the A1 and A2 are electrically connected. In that one LinkedIn article diagram, he was showing the artifact only on one channel, one of the temporals T3 or T4. Linked ears is especially helpful on the temporal sites as the reference is so close to what is being measured.
  • knightv8knightv8
    I understand. I will make an assumption in my paper that clinical equipment might be more targeted to capture this artefact; at least from my findings. I stand to be corrected...

    The artefact was showing only on T3 and T4 since that was on bipolar montage. The diagram below that one, is the ipsilateral (referential to A1) which shows the artefact on all electrodes which are referenced to A1.

    Any ways, many thanks for your feedback mate; I thought i'd better double check with fellow researchers :)

    Patrick
  • wjcroftwjcroft Mount Shasta, CA
    His ipsilateral chart is showing artifact of about 80 uV (all 12 channels using A1 reference). That seems remarkable and very unusual. All clinical neurofeedback / QEEG equipment uses linked ears reference.

    The Texas Instruments ADS1299 A/D converter used in OpenBCI, IS used in some "clinical equipment." I've seen a couple of commercial amps with specs almost identical to the TI spec (although they don't reveal that engineering detail of their internal circuitry). Implying that they use the ADS1299.

    Regards,

    William
  • knightv8knightv8
    that was my thought exactly... how was he able to get a 80uV on all A1 referenced channels...

    Did not know re:ADS1299 used in clinical equipment. Thanks.

    They must have some internal circuitry different then ours to be able to capture that signal with such strength and I am not even able to replicate it once.

    Thanks again mate

    Patrick
Sign In or Register to comment.

© Vanilla Keystone Theme 2026