simtk.org

I would like to thank you for developing a great toolbox. I have a few queries that I would really appreciate it if you be able to answer please.

1) One of the input quantities required for running CEINMS simulation is muscle excitations which are simply the normalized linear envelope of EMG experimental data. In CEINMS manual, it has been suggested that the MVC of each muscle to be used for normalization of EMG signal. However, MOtoNMS uses the maximum peak from each muscle of all trials (if multiple trials were chosen for elaboration) to normalize the data. I would like to know if this is a correct procedure in general? would it be alright to proceed with maximum peak defined for each muscle as a normalization technique? if no, is it possible to implement MVC data in MOtoNMS?

1.1) In addition would it be acceptable if normalisation has been done trial by trial or normalisation must be done all together based on the single highest value for each muscle?

1.2) in case of 1.1, In ElaborationInterface, in the section where we need to 'select trials to Max EMG', If we select all the trials, will this find the maximum value for each muscle from all trials and divide that value to the corresponding muscle for each trial? would that be the correct procedure?


2) In the dynamicElaboration, the EMG result has been plotted for different muscles (AllNormalizedEnvelopes) in which Y axis has been labelled as normalized Envelope (%max). Thereby as an example, if 6 trials are included in dynamicElaboration, would the AllNormalizedEnvelopes plot be the average of all the signals for each individual muscle from all trials??

2.1) In addition, in the plot of an individual muscle, the amplitude (micro volt) is normalised to Maximum peak and represented as percentage. If the signal represents muscle excitation, Do 0% and 100% mean that muscle is not active and fully active, respectively? and what does it mean if the peak of the signal show 20% (e.g. y axis for rectus femoris is 20%)?

3) Is there an option in MOtoNMS to calculate muscle activation from muscle excitation?

Dear Pouyan Mehryar,

thank you for your appreciation for our toolbox.
In the following are the answers to your questions:

1)
I'm pleased to inform you that MOtoNMS does not use the maximum EMG peak from each muscle of all trials to normalize the data. Instead, you can easily refer to MOtoNMS User Manual (http://rehabenggroup.github.io/MOtoNMS/ ... processing) to find the actual procedure:

Maximum value for each EMG signal is estimated from one or more trials defined by the user: when available, a trial of maximal voluntary isometric contraction can be used, as well as a subset of the input trials.

We are currently dealing with problems related to EMG normalization (pag 35 of this book of abstract: http://iscsb2015.lboro.ac.uk/ISCSB_book ... tracts.pdf), and one of the main new features of Release 2.2 is indeed the improvement of the step related to the processing of EMG signals (http://rehabenggroup.github.io/MOtoNMS/ ... tures.html).

1.1) and 1.2)
In MOtoNMS, the maximum value for each EMG signal is computed as described in 1), and this value is used to normalized the corresponding EMG signal in all the trials of interest.
If you select all trials, all trials will be used to compute the maximum EMG value for each muscle.

2)
I suggest you to look at the results you get in the output folders. I think the answer should be very intuitive. Indeed, a separate folder is automaticaly created for each trial selected for the processing. The AllNormalizedEnvelopes figure is stored for each trial, and therefore it includes all normalized envelopes of that specific trial (resulting from all the collected EMG signals). This means they are not averaged.

2.1)
Those plots represent the normalized EMG linear envelope for each individual muscle (i.e. high pass filtered (30-300Hz), full wave rectified, low pass filtered (6Hz), as in Lloyd and Besier 2003).
Maximum EMG values, and consequently normalized amplitudes, depends on the trials selected by the user for this purpose.
About the Y axis, only the scale is adjusted for a better visualization.

3)
No, there isn't. MOtoNMS was not conceived as an implementation of a neuromusculoskeletal software, and we intentionally kept the steps needed to perform forward simulations separate. MOtoNMS represents just the first step towards the use of other neuro-musculoskeletal software, such as OpenSim and CEINMS, as it processes data from different acquisition systems and provides input for neuro-musculoskeletal software. For this reason, it only implements the preliminary EMG processing steps required to obtain muscle activations.

Hope this helps.
Best regards,
Alice Mantoan