Tweet
Hi, everyone, I want to ask if you know the normalization method used in EMG data, they are Maximal Voluntary Contraction (MVC), peak dynamic method(PDM) and mean dynamic method(MDM). Do you know the difference between them? I have done some literature review and only know MVC is may not represent the maximum activation capacity among patients, as a result, it may not valid in clinical. In addition, there is little difference between PDM an MDM. Thank you in advance!
-
-
Re: the difference between normalization of EMG data
There is a lot of literature on this topic. It seems to me that the answer like many things is it depends on what purpose you are scaling the EMG for.
If the purpose is to reduce variance between subjects then I don't think the protocol matters that much, as long as all subjects are doing the same thing as the reference muscle action.
If the purpose is to define the EMG over a range that represents all feasible values from "off" to "as high as possible" (e.g. for input to a muscle model) then that is a much more difficult problem. I think it's clear, as you noted, that MVICs do not do that well. I like to use some "max effort" task like a short sprint or a jump for maximum height, but that only works well for some muscles or for subjects who can safely do those motions.
Hope this helps,
Ross -
Re: the difference between normalization of EMG data
thank you so much, RossComment
-
Re: the difference between normalization of EMG data
Essentially "Maximal Voluntary Contraction" is just that - it's the best contraction that you can get with the subjects' cooperation - and it's often not as large a contraction (in terms of EMG signal levels) as the subject will generate in an active physical situation. This doesn't invalidate it as a method of comparison for a subject, but I think that it does mean that comparisons between subjects need to control all the external factors, I would not be surprised if two different investigators generated different MVC levels from the same subject om the separate tests.
Another factor to look out for in MVC testing is that you must ensure that the recorded EMG levels in the MVC test and the active trial use the same signal gain and the gain must be low enough so that the EMG signal levels do not clip or exceed the dynamic range of the EMG system and the recording system. If the EMG signal is clipped at any point during the active trial or the MVC trial then the data is invalid.
Comment
-
Re: the difference between normalization of EMG data
I would recommend using sub-maximal voluntary isometric contraction to normalize the EMG signal. (see - Burden https://doi.org/10.1016/j.jelekin.2010.07.004)
We have used the ability of the patient to maintain an anatomical segment against the force of gravity as submaximal task (see -Tabard et al. https://doi.org/10.1016/j.gaitpost.2017.10.026)
Hope this helps,
Stephane ARMANDComment
Comment