I've found many papers describing different methods for processing but none for clinical interpretation or how to perform the test properly to get fatigue indicators from power spectrum or other signals.
Hi,
I go along with your view that few papers have addressed the physiological interpretation of signal processing techniques. However, you can find some explanations for the conventional approaches such as root mean square (RMS), mean frequency (MNF) and median frequency (MDF), as the most widely used variables in EMG studies related to fatigue, in the literature. Here is an article, from which I got an insight into underlying processes of fatigue. Hope it is advantageous for you as well.
Yours,
Fatigue is notoriously difficult to define for individuals. I am not sure if there are any concrete clinical indicators of fatigue based on EMG. The best detection methods I am aware of use wavelet analysis of EMG data, looking for frequency shifts to indicate fatigue in the muscle.
I would recommend the use of IIS index (refer Kumar et al, IEEE TNSRE, 2011). This has been tested along with the biomarkers. Have a check, and let me know how it goes.
Myoelectric fatigue, measured from surface EMG (sEMG), is an old matter: MNF (and MDF) slope decrease is a typical sign of spectral compression toward lower frequencies (i.e., type II greater MU activation). See the book MERLETTI R., PARKER P. “Electromyography: Physiology, Engineering and non-invasive applications” IEEE Press, Piscataway, N.J. (USA), 2004, pp. 281-305
Our work suggests that spectrum is not a good measure, and only the IIS can be used. Check IEEE Trans Neural Syst Rehabil Eng. 2011 Oct;19(5):578-87. doi: 10.1109/TNSRE.2011.2163527. Epub 2011 Aug 15.
Measuring increase in synchronization to identify muscle endurance limit.
- Badges
- Science topic