More specifically, is it possible to calculate the signal-to-noise ratio (SNR) for EMG signals? For example I thought of computing the ratio between 100% MVC and 0% MVC amplitude (baseline noise) but I'm not sure whether this is a correct approach as noise may increase with higher signal amplitudes as well.
This is an excellent question. There is no standard method to evaluate quality of EMG. I have been doing some research to find an automated method to validate EMG (ensure it is of adequate quality to use), and to detect and identify contaminants. You can refer to the paper Automated Biosignal Quality Analysis for Electromyography Using a One-Class Support Vector Machine, which includes background on this very topic and reviews what existing literature we could find on EMG quality assessment. It also refers to some targeted work that we did for particular contaminants (e.g., ECG, motion artifact, clipping, power line interference).
If you are looking for a validation method, in the paper A novel approach to surface electromyography: An exploratory study of electrode-pair selection based on signal characteristics, we looked at an electrode array and tried to find a metric that could identify electrode pairs that yielded acceptable EMG quality.
http://www.delsys.com/educational-resources/knowledge-center/tutorials/
Noise may not depend on the EMG amplitude, therefore you can calculate SNR by means of the baseline noise.
"The actual quality of the EMG data is often referred to in terms of the signal-to-noise
ratio (SNR), which is given in decibels, so it is the ratio of signal power (PS) to noise
power (PN).
...
SNR=10log(PS/PN);" - "Kamen and Gabriel, Essentials of Electromyography"
This is an excellent question. There is no standard method to evaluate quality of EMG. I have been doing some research to find an automated method to validate EMG (ensure it is of adequate quality to use), and to detect and identify contaminants. You can refer to the paper Automated Biosignal Quality Analysis for Electromyography Using a One-Class Support Vector Machine, which includes background on this very topic and reviews what existing literature we could find on EMG quality assessment. It also refers to some targeted work that we did for particular contaminants (e.g., ECG, motion artifact, clipping, power line interference).
If you are looking for a validation method, in the paper A novel approach to surface electromyography: An exploratory study of electrode-pair selection based on signal characteristics, we looked at an electrode array and tried to find a metric that could identify electrode pairs that yielded acceptable EMG quality.
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Signal Processing