James, are you only interested in determining the global mean/median frequency of the EMG signal? In this case, one method is to simply use FFT to transform the signal to frequency space and then take a mean or median of the content.

you should separate the periods in which the muscle is actually active (in order to avoid the presence of any kind of background noisy activity which could negatively bias your spectral measurements), then you can calculate the power spectrum (you can use the square of the FFT, or some parametric spectral estimation technique). You can calculate the global mean/median frequency for each set, or determine it for each repetition burst by burst, if you are interested in some kind of temporal evolution of the spectral properties. Also Short Time Fourier Transform can be suitable.

I agree with Cristiano Marchis: spanning through the range of motion of your movement the electrodes receive influence from different portions of the muscle (that is sliding under the skin), thus it would make sense to perform an angle-based appropriate windowing (assuming a reasonable amount of sliding velocity of the muscle) and perform spectral or amplitude analyses taking into account that you can only compare windows from the same angular position. The joint angle information, or any other geometrical reference, is however needed.

You can take a look at the following reference. It permits the computation of mean frequency during dnamic tasks (as in your case) relying on covariance function .

Conforto S, D'Alessio T, Real time monitoring of muscular fatigue from dynamic surface myoelectric signals using a complex covariance approach, Med Eng Phys 1999.

I recommed one paper. This paper is the paper considering consecutive contraction like you. However, they performed during consecutive concentric contractoin not eccentric at biceps brachii muscle. I think this paper will help you. [REFERENCE] Beck, T.W., Housh, T.J., Johnson, G.O., Weir, J.P., Cramer, J.T., Coburn, J.W., Malek, M.H., 2005. Comparison of Fourier and wavelet transform procedures for examining the mechanomyographic and electromyographic frequency domain responses during fatiguing isokinetic muscle actions of the biceps brachii. Journal of Electromyography and Kinesiology 15, 190–199.

I aim to look at differences between the rectus femoris, vastus lat and vastus med. Have you much experience using a Bagnoli Delsys system and EMGWorks analysis?

James,

the issue of EMG spectral estimates during dynamic muscle actions is not a simple one. As Leonardo Gizzi already answered, relative movements between electrodes and muscle must be taken into account and it is absolutely not appropriate to compute a global burst MEAN/MEDIAN frequency value because of this. I attach for your convenience an excellent review on this topic by Dario Farina which appeared on Exerc Sp Sci Rev in 2006.

Besides, you can also download from my RG pages the following papers.

My best

Non-invasive assessment of muscle fiber conduction velocity during an incremental maximal cycling test. Sbriccoli et al, JEK, 2009

Exercise induced muscle damage and recovery assessed by means of linear and non-linear sEMG analysis and ultrasonography. Sbriccoli et al, JEK, 2001

Surface EMG modifications after eccentric exercise. Felici et al,JEK, 1997

James, here is the file

Hey Sean, I'm looking to determine global frequency. I aim to calculate for each individual contraction, but establishing set means. Thanks for your suggestions, that've been really helpful.

Drs. Gizzi and Felici are correct in stating that the use of median frequency is most likely inappropriate to assess the frequency spectrum during dynamic contractions. The movement of muscle fibers beneath the skin over which the surface electrode sits results in the recording of electrical activity from different motor units throughout the dynamic contraction. However, a method has been proposed for estimating the time-frequency parameters in cyclic dynamic contractions from surface EMG. I suggest you consider the paper by Bonato et al. (2002: Time-Frequency Parameters of the Surface Myoelectric Signal for Assessing Muscle Fatigue during Cyclic Dynamic Contractions). Perhaps this methodology would enhance the validity of your findings.

Attached is the above listed paper.