Electromyography (Muscle sensing)

[Kim O'Sullivan]

Hi guys,

I was wondering if anyone has had experience with sensing muscle contractions, or the electrical impulses that cause them? 

I've seen a few different articles on Google about it, but I can't find any established method of doing it, just a bunch of different university project-type discussions and papers about it in various ways.

Any ideas? I'm looking for something with a reasonable amount of resolution, not just contracted/relaxed. It's for a biofeedback system, to train people in the correct usage or avoidance of muscles during certain movement exercises.

Cheers,

Kim

[Matthieu Boardman]

Cliff Eddens and myself (Matt Boardman) did the buglab diy ekg class this last summer and actually had to work around electromyographic signals in our attmept to accurately measure the electrocardiographic signal we were after.  

However, that did give us the idea to do a follow up DIY EMG class and then maybe go for the gold with an EEG class later on.  While I am no expert on EMG analysis I can tell you that to get the accuracy you're looking for you will likely be wanting to do some high pass filtering as the EMG signals tend to be small in amplitude and high in frequency...just like many sources of EMI.  Also the placement and size of the electrodes will play an important role in your specificity.  

I would love to work with you on this project as I am moving in this direction regardless.

Anyone else have any more specific EMG experience out there?

-Matthieu

--
The truth does not change according to our ability to stomach it.
 - Flannery O'Connor

[Carlos Chiossone]

Does this have anything to do with also reading regions of the brain at different stages? I am working with Princeton U on a project on hypnosis and brain stages which use a machine to measure the levels. It is not only read from the brain but arms and legs. I have fairly good access to the machine and am supposed to be trained on it next month.

 

c

[Matthieu Boardman]

As far as I know the brainwave stuff is all electoencephalography
(eeg). This is just a guess but the arm and leg leads are likely
there to detect (and gate out) any ekg signals from your eeg. I think
the two signals are both low enough in frequency that they cannot be
separated using filtering alone...this is just a guess though.

-M

[raphael]

I did some of that just recently. I used an instrumentation amplifier (3
op amps in a special configuration, look it up in wikipedia). It's
pretty hard to do, but I got some data.

Also look at ECG projects, it's more or less the same circuit.

--
Raphael Abrams
Artist Engineer

web: teuthis.com
drawings: doodlebird.wordpress.com
hack space: nycresistor.com
email: rap...@teuthis.com

[Ben Simpson]

It's all about the sensor.. acquire your analog signal using the sensors (you probably found or will find on google), use an instrumental amplifier to gain it up then filter, etc. Remember filtering can be done digitally too.

[Tom Richner]

Don't forget electrocorticography (ECoG), but it does require a 10 cm craniotomy.  Its like EEG, but the electrodes are put on surface of the brain. 

More seriously, I'll echo the instrumentation op-amp suggestion for either EMG or EEG.  I recommend using an integrated circuit with all 3 op-amps on one chip.  You can get chips where a single external resistor sets the gain of the amp.  However, if you want to reduce environmental noise, you'll want to put the voltage follower op-amps of the instrumentaiton amp within a cm of the electrodes, in which case an IC would not be the best.

on limb electrodes with EEG:
EEG is very sensitive to the slightest movement of the subject.  Putting electrodes elsewhere on the body can help detect movements, so that the signal can be corrected with independent component analysis (ICA), or the data during movements can be simply tossed out.  Alternatively, muscle tone can very with depth of sleep, so it would be useful to record EMG in addition to EEG when studying levels of sleep and hypnosis.

Can some one suggest an affordable analog to digital converter with multiple channels?