I was wondering whether you might share your experience with skin preparation for facial EMG (measurement of facial expression of emotions). My biggest concerns are: (1) is it necessary to abrade the skin, and (2) how do you practically handle bearded participants? Do you shave them or ask them to shave before the experiment, e.g. during recruitment? I would also appreciate any other tips that will help us run our first facial EMG experiment.
Hi Lukasz,
Prior to electrodes placement, you must properly prepare the participants’ skin in accordance with SENIAM recommendations (Hermens, Freriks, Disselhorst-Klug and Rau 2000): the skin is rubbed with gauze soaked in alcohol and a conductive gel need to be applied to the electrodes to reduce the impedance of skin-electrode interface below 5 kΩ and thus improve the signal-to-noise ratio.
For bearded participants, there are two solutions: 1) beard is short and you can use the procedure described above; 2) beard is too long and you will have to shave it before skin preparation. As you said, it is necessary to inform participants during recruitment.
I hope it was helpful for you.
Best
Mathieu
Hi Lukasz,
Prior to electrodes placement, you must properly prepare the participants’ skin in accordance with SENIAM recommendations (Hermens, Freriks, Disselhorst-Klug and Rau 2000): the skin is rubbed with gauze soaked in alcohol and a conductive gel need to be applied to the electrodes to reduce the impedance of skin-electrode interface below 5 kΩ and thus improve the signal-to-noise ratio.
For bearded participants, there are two solutions: 1) beard is short and you can use the procedure described above; 2) beard is too long and you will have to shave it before skin preparation. As you said, it is necessary to inform participants during recruitment.
I hope it was helpful for you.
Best
Mathieu
Dear Dr. Kaczmarek,
Practically, appropriate skin preparation is indispensable for a good electrode-skin contact bringing about better EMG recordings, fewer and smaller artifacts and noise, as well as lower risk of imbalance between electrodes. It is, thus, conceivable to ask the participants to shave before recordings. For example, in the written informed consent or in the pre-examination instructions form, the subjects can be informed of necessity of shaving. From my point of view, it is preferable to do it yourself before the commencement of examination. However, if I were you, I'd exclude the bearded participants with higher preference to recruit females to ensure the high-quality recordings.
Anyway, another potential problem may be the carotid artery branches as well as other arteries activity in facial region contaminating EMG recordings with ECG artifacts. Then, there may be a need to record ECG signals separately, if you are willing to attenuate such artifacts using adaptive noise cancelers.
In addition, you may record EMG signals either bilaterally or unilaterally. In the latter case, the site of recording seems to be important. If the subjects are all healthy, then, considering the hand dominancy is a reasonable idea.
Regards,
Nice question Lukasz!
Certainly facial skin is more sensitive than other areas of the body and alcohol cannot be used excessively or it might cause abrasion.
Try cleansing with dry soap that leaves no oil residue.
The person has to shave.
Use electrolyte gel but not in excess so that the electrodes would still adhere to the skin.
Dear Lukaszi
What facial muscles are you planning to record. The preperation for the near the chin, and on the forehead will be very different. The bigger concern should be- how will you get the correct inter-electrode distance for differential recording. That may lead to bigger changes in the signal properties.
What is your application?
Dear Dinesh, we target corrugator supercilii (frowning) and zygomaticus major (smiling). Very much the same way as in this picture: http://ucfretrolab.org/wp-content/uploads/2012/07/EMGAttached.jpg
Thanky you and also Zaid, Fariba, and Matheiu for your interest and help.
dear Lukasz
We have a few small papers on the use of facial EMG. Have a look at these papers, and see if the information is sufficient. Google for- Kumar, voiceless speech. Also, Kumar, facial EMG. Some of these are:
Sridhar Poosapadi Arjunan, Hans Weghorn, Dinesh Kant Kumar: Reliability and variability in facial electromyography for identification of speech and for human computer control: an experimental study. IJMEI 3(4): 311-330 (2011)
Conf Proc IEEE Eng Med Biol Soc. 2006;1:2191-4.
Unspoken vowel recognition using facial electromyogram.
Arjunan SP1, Kumar DK, Yau WC, Weghorn H.
Effect of Noise on Facial EMG
Sanjay Kumar*(1), Dinesh Kant Kumar (2), Melaku Alemu (1), and Mark Burry (1)
I am sure that the papers are available on-line, but if not, let me know and I will arrange for the same.
Dinesh
After seeing the photo, my guess is that your work is similar to the last reference- Effect of Noise on Facial EMG
Sanjay Kumar*(1), Dinesh Kant Kumar (2), Melaku Alemu (1), and Mark Burry (1) , which appeared in Acoustics, 2004.
When you record surface EMG (sEMG) - as with all other bioelectric signals, such as ECG, EEG, ERG, EOG, EGG, and so on - you MUST abrade the skin in order to reduce the number of dead skin cells, to eliminate skin fat. Obviously, dealing with facial EMG recordings you must operate with high carefullness because face skin is particularly delicate. For males, depending on electrodes positioning, the subject must shave (at least locally where the electrode NEED to be posed
- Badges
- Science topic
- Similar topics
- Psychology
- Emotion