Dry electrode Raw Ecg Signal Processing

Hi. We aim to receive ECG signals with dry electrodes. We are investigating which filter will be better for us if we apply it with MATLAB. The signal is noisy and there is fluctuation. what could be the reason?

Our electronic circuit has a 0.5 hz high pass and a 40 hz low pass filter.

Answer 1

[Star Strider]

Without knowing the experimental set-up, it’s not possible to say. A pair of electrodes connected to an amplifier are prone to any number of problems, from the electrodes themselves being capacitors (differentiators or highpass filters, and this also depends on their construction) to simply picking up random noise, movement artifacts (the electrodes themselves may not be in close contact with the skin surface), or any number of possible sources of interference. A third reference electrode (normally designated ‘right leg’) with the appropriate configuration of differential amplifiers should eliminate some of the observed noise, as well as some low-frequency motion artifact.

There are several papers on dry electrodes that I was able to pull up in the literature (most of which are behind paywalls), so I suggest that you search PubMed for relevant articles, if you have not already.

.

[Cansu Baser]

We used stainless steel electrodes. and we added it on the right leg electrode. Filters pass a passive high and 5. Butterworth is a low-pass filter. Only the distances of the electrodes are very close to each other, and would it be useful if we added a 50 Hz notch filter?

Thanks for the advice. We've done a literature review, but there may be different articles that we can access.

[Star Strider]

With respect to the notch filter, the reference electrode should prevent powerline interference, since the reference electrode potential should be subtracted from the other leads with a differential amplifier. I would only include a 50 Hz notch filter if the Fourier ttransform of the signal demonstrated a significant 50 Hz spike.

Also, the leads should be shielded (coaxial) with the centre llead providing the signal, the shield grounded to the instrumentation common ground, and the instrumentation needs to have its common ground connected to the earth ground. That should prevent much of the noise.

.

[Cansu Baser]

I added the schematic and the rough drawing of the circuit board. Thank you. I think there might be a problem with the ground line. There's a surge in the signal.

[Star Strider]

I can barely read the schematic (I sould have to research the chips and their connections to make sense of it anyway), and I don’t understand the second figure.

It is always necessary to use coaxial leads with grounded shields. Ground faults (differences in potential between grounding points) are always a problem, and the reason for the ‘plug panels’ in ICUs, to keep the ground distances short and ground faults to a minimum.