IAA with MOCO results

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
David Torres
Posts: 17
Joined: Wed Feb 11, 2015 6:38 am

IAA with MOCO results

Post by David Torres » Tue Mar 14, 2023 4:42 pm

Hi everyone
I've been trying to apply IAA to MOCO optimization results to get muscle contributions. Particulary, I used Ross Miller's Model (UMocoD) and results included at same project. I set up IAA on Opensim GUI using several setup files as guide example, however, the results for each muscle aren't consistent.
For instance, if i compare my results with same analysis results of Rajagopal's model, using same IAA setup, they dont look similar.

GAS and RECFEM contribution. Rajagopal's model
Image
GAS and SOL contribution. Miller's model
Image

I have tried for several ways such that deleting contact model, changing IAA setup, editing model, but I always receive weird results.
What could it be ? If someone could tell me what's is wrong or give an advice, I would be grateful.
Thanks in advance
Attachments
Miller model.rar
(227.36 KiB) Downloaded 13 times
Last edited by David Torres on Wed Mar 15, 2023 11:59 am, edited 1 time in total.
Tags:
Windows
Top
User avatar
Nicholas Bianco
Posts: 1050
Joined: Thu Oct 04, 2012 8:09 pm

Re: IAA with MOCO results

Post by Nicholas Bianco » Tue Mar 14, 2023 11:32 pm

Hi David,

The results of IAA are pretty sensitive to the type of model you choose. How are you modeling foot-ground contact? In the Hamner et al. 2010, they needed to use a rolling constraint in order to produce realistic contact forces. It's worth checking out that first paper if you want to use IAA: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2973845/.

Best,
Nick
Top
User avatar
David Torres
Posts: 17
Joined: Wed Feb 11, 2015 6:38 am

Re: IAA with MOCO results

Post by David Torres » Wed Mar 15, 2023 10:27 am

Hi nick
Thanks for your answer.
How are you modeling foot-ground contact?
I used RollingOnSurface constraint to model foot-ground contact. I made use of another hamner's project to configure my own analysis.
Here is my IAA setup
Image
The results of IAA are pretty sensitive to the type of model you choose.
I had already checked out hamner et al. work but I didnt noticed something related to sensitive. I'm going to ckeck if I can fix the problem. However I would like to known if there is another way compute muscle contribution .

Thanks in advance
Best,
David
Top
User avatar
Nicholas Bianco
Posts: 1050
Joined: Thu Oct 04, 2012 8:09 pm

Re: IAA with MOCO results

Post by Nicholas Bianco » Wed Mar 15, 2023 11:20 am

Hi David,

It seems like the soleus and recfem contributions are exactly the same in the Miller model, and the curve looks like the ankle net joint moment, so double check that you're not plotting the soleus information for the recfem.

The plots from the Rajagopal model suggest that there's something wrong with the ground contact constraint.

Keeping digging into it and let me know what you find.

Best,
Nick
Top
POST REPLY