segment power analysis

[David Torres]

Hi everyone
i have been working on transtibial amputee gait and I was able to run some MOCO simulations getting some good results. I would want make a segment power analysis but I dont know how. I have tried using IAA tool, however, I got weird results.
Than you for any help!
David

[Nicholas Bianco]

Hi David,

Could you be a little more specific? Do you want to know the power produced at each joint in the model, or something else?

The "analyze()" utility, which you can find in the Moco example, is a great way to compute information from a Model given a MocoSolution. I encourage you to try that approach first.

Best,
Nick

[David Torres]

Hi Nicolas.
Sorry for not being clear. Particularly, I want to make a analysis just like picture below.

This picture was obtained from paper titled "Altering prosthetic foot stiffness influences foot and muscle function during below-knee amputee walking: A modeling and simulation analysis"

I would like to show muscles and foot prosthetic contribution on stance phase of gait.

i hope I made clear this time
Thanks for your attention
David

[Nicholas Bianco]

Hi David,

I'm not sure how this paper performed this analysis, so I can't tell if the analyze() utility is sufficient, or if something like IAA is required. If you could provide more details on exactly what you're trying to do and where you're getting stuck with Moco tools, I can be more helpful.

-Nick

[David Torres]

Hi nick

If you could provide more details on exactly what you're trying to do and where you're getting stuck with Moco tools, I can be more helpful.

Ok NIck. I will try to do my best effort. In my research I want to found how stiffness of foot prosthetic influence on ampute gait. I modified muscleskeletal model deleting part of limb and adding transtibial prosthesis. Aforementioned prosthesis has a multibody foot prosthetic model, whose objective is to depicte viscoelastic features of it. Finally, I set optimal control Problem using MOCO to know what happens if foot prosthetic is compliant or stiff.

Fey and Neptune (Altering prosthetic foot stiffness influences foot and muscle function during below-knee amputee walking: A modeling and simulation analysis) made something similar but using dynamic optimization to track experimental data. They used GRF descomposition and segment power analysis to show how muscles and foot contribute to propulsion of center of mass. I tried making similar analysis using IAA tool but it didn't work. I would like to know if there is another way to show muscles and foot contributions or how it should make aforementioned analyses using MOCO/Opensim tools

Thank you Nick for your patience.
Regards
David

[Nicholas Bianco]

They used GRF descomposition and segment power analysis to show how muscles and foot contribute to propulsion of center of mass.

I'm not familiar with these methods. Let me know what you need to calculate from the simulation results and I can point you to the right Moco tools.

Best,
Nick

[David Torres]

Hi Nick
I was reviewing and it would be helpful how to get dynmics information such as centripetal and centrifugal forces, gravity forces, mass matrix from Moco solution. I could use this data for making my own analysis.

Thank you again for you patience
Regards
David

[Nicholas Bianco]

Hi David,

After briefly review that paper, it seems like an induced acceleration analysis is really what you need for this. You could probably recreate what the IAA tool does for your segment power analyses, but that would require accessing the SimbodyMatterSubsystem from the model

Code: Select all

matter = model.getMatterSubsystem();

and calculating the mass matrix (and whatever else you need) manually

Code: Select all

matter.calcM(state, M);

Simbody doesn't form the mass matrix, and other common dynamics structures, until you explicitly ask for them since it is based on a Featherstone-like algorithm. You can get a states trajectory from an MocoTrajectory via exportToStatesTrajectory().

That said, I would recommend trying to sort out your issue with the IAA tool before trying to create your own method. Unless you are sure the IAA tool won't work for your study, it will very likely be faster to fix your issue with IAA than build a new tool. For help with IAA, please post your questions the main OpenSim forum.

Best,
Nick

[Carlos Gonçalves]

Hello Nick and David,

I'm also interested in using IAA with my result from OpenSim Moco.

From what I researched, the IAA will use External Loads and Kinematic Constraints in the foot to solve the problem. Wouldn't it work with the Contact Forces in the feet?

I already had the Analyze tool working with ForceReporter, Body Kinematics, Point Kinematics and Mucle Analyze just passing the control.sto and state.sto files. Should I emulate the IAA workflow creating my own "External Load" file from the contact forces?

Best regards.

[Nicholas Bianco]

Hi Carlos,

From what I researched, the IAA will use External Loads and Kinematic Constraints in the foot to solve the problem. Wouldn't it work with the Contact Forces in the feet?

Yes, since IAA will require some sort of kinematic constraint at the foot, I'm not sure what is the best way to convert Moco simulation results to IAA. It would probably be best to review the workflow from previous IAA studies first (Sam Hamner's study is on SimTK), and then think about how to achieve that with a Moco solution.

Best,
Nick