0
\$\begingroup\$

I'm working on building an acquisition system for sEMG signals. The problem I'm facing is how can I reliably measure the performance of different designs? Obviously I can just put the electrodes to my arm and see the signals on the oscilloscope, but that's not very objective way. The quality of signal can be affected by things like:

  • position of the electrode on the muscle
  • amount of sweat on the skin
  • how strongly I contract the muscles
  • muscle fatigue

I would like to have some sort of artificial muscle that I could attach the electrodes to, and that would always generate the same, repeatable signal. How can I do it?

For testing stuff like filtering software and hardware I can solder wires in place of electrode plates and connect them to signal generator, but I would like to do the end-to-end testing of the complete system.

Picture of electrode for reference: Picture of electrode for reference

\$\endgroup\$

1 Answer 1

1
\$\begingroup\$

As you are already aware, that dry-contact electrode for surface EMG is subject to many variables -- skin prep, skin impedance, hair, sweat, positioning, effort, patient anatomy and more.

It would be challenging to build an accurate anatomical simulator including all these variables -- probably a PhD dissertation's worth of work.

Instead, I'd suggest a short cut: "A/B Testing". Using yourself, or a volunteer, as the subject, compare two competing electrode designs. Install them as close together as possible over the muscle area of interest. Record the data from both simultaneously, and compare for signal strength (or SNR). Then, switch their positions and repeat.

If you have many possible designs, follow a binary tree (like a sports tournament) to determine the winning design.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.