FreePostureFullBody - Model

Hi everybody,

for a current project I intend to study the muscle forces of the arm-shoulder-system for different arm movements and handled weights. For this purpose I would like to work with the “FreePosture” Model. But in the model description is mentioned, the model should only be used for kinematic analysis. In a first run I have already performed an inverse dynamics analysis and got good results. Would you suggest to work with another model e.g. standing model or is is the FreePosture model for this kind of analysis valid as well?

Thanks a lot for your comments!

Greetings
Stefan

Hi Stefan,

the main difference between the FreePosture and the StandingModel is in the attachment of the human body to the rest of the world: The FreePosture is fixed to the ground at a node in the hip segment, the StandingModel is attached at the feet as the name intends.

If you are thinking of switching the model you are starting from my suggestion would be to start from the Human template model (available in from version 6.0 onwards). This is the clean implementation of a human model similar to the FreePosture model. If you are only interested in arm movements you can easily switch of the legs and maybe also one of the arms using the BM statements to make the model faster.

Best regards
Daniel

Hi Daniel,

thank you very much for your quick response!

My aim is to examine the movement on the upper extremities for sitting postures. To model the movements I have to use the interpolation function (as already implemented in FreePosture). From my point of view I have to spend a lot effort to integrate the interpoliton function to the full body model you mentioned?!

Regarding my research focus, I would suggest the FreePosture model with the fixation at the hip should be appropriate as well? Am I right?
In my opinion in that case the inverse dynamics analysis should provide valid results. What do you think?

Best regards,
Stefan

Hi Stefan,

The reason why it is said the the model is only appropriate for kinematics is because it does not contain muscles. I think when you switch on the muscles in the FreePosture model it should work as well for the inverse dynamics, especially if you already saw good results.

Best regards
Daniel

Hi Daniel,

thanks again for your support!

Best regards,
Stefan

Hey Daniel,

I have got one more question concerning the Body model.
In the FreePostureMove model and the Standing model I have applied exactly the same movement of the right arm (with a weight in the hand (vertical forces (each 1,5N) appear on 27 nodes in the hand)).

In my understanding the results (e.g. muscle activity) of the simulation should be almost similar.

But I get significant differences e.g. in the level of muscle activity, especially for shoulder. What is the reason for this effect? The muscle models should be the same in both models (simple muscle model)?!
According to my analysis the Standing model is weaker than the FreePostureMove model??

Thanks for your comments!

Regards,
Stefan

Hi Stefan,

It is difficult to say where the difference comes from. In principle the models should be the same, and as far as I see there is no scaling in any of the model.

I tried something similar starting from template models, one using the standing human and the other using the human, and applied in both models a GH abduction with constant speed. When I looked at the Deltoid muscles, I got the same results in both models.

Are you sure that everything else is the same in both models (motion, gravity, loads, both using the same scaling specified in the main file, …)?

Best regards
Daniel

Hi Daniel,

attached to this post you can find the mannequin files of StandingModel and FreePostureMove model where I have applied exactly the same movements and forces.
In the separate Excel file are the summarized results of the simulations. Especially the muscle activation of the shoulder muscles differs significantly for several muscles. Where does this discrepancies come from?

However I have made no changes in gravity and scalings at all.

Thanks for your support!

Regards,
Stefan

Hi Stefan,

I am trying to run your model. Which version of AnyBody and which AMMR are you using?

Best regards
Daniel

Hi Daniel,

I am using the latest version of AnyBody (6. 0. 1. 3928 (64-bit version)) and AMMR ( Version 1.6.1).

regards,
Stefan

Hi Stefan,

I think I could load and run the models. The differences I can see is that there is a difference in the velocities of the model. If you increase the number of time steps in both models e.g. to 10, you even see that the hand of the standing model is moving whereas the hand free posture model is not. That might explain the differences in the forces.
You could try to make sure that the velocities of drivers are the same.

Best regards
Daniel

Hi Daniel,

thanks for your advices.
I try to follow your instructions but unfortunately I could not realise these. I do not understand how I can set up the same velocity for both models.

Especially I am not quiet sure about the impact of the “nsteps” on the velocity. How is the duration of my study defined and linked to the velocity?

Thanks for your support.

Regards,
Stefan

Hi Stefan,

In general the nSteps variable should not have an influence on the velocity since the velocity is defined in the drivers of the model, at least as long as nSteps or tStart, tEnd or another variable defining the simulation duration is not used in the velocity definition of the drivers.
The duration is given by tStart and tEnd (or alternatively tArray), you can find the values of these variables in the model tree in the study if not explicitly defined in the code.

Best regards
Daniel

Hi Daniel,

thanks a lot for your help!

Regards,
Stefan