simtk.org

Hi all!

I'm starting a new study which involves the use of OpenSim.
I'm working in a gait lab which strongly believes in the anatomical calibration approach introduced by Capozzo et. al and that has not previous experiences with OpenSim.
I haven't seen any study yet with a similar markerset and protocol in OpenSim, and I was wondering if someone has already tried it.
I would like to understand if the markerset used may influence scaling and/or also IK results.
If yes, may I say that a protocol or markerset is "more suitable" than another for OpenSim?

Thank you very much for your help in advance.
I will be very grateful for any suggestions and observations.

Alice

Hi Alice,

A range of marker sets are possible, with IK allowing fewer markers than that required in segment based modelling but also permitting redundant marker sets where IK solves an over determined problem.

For both scaling and IK, I would say the issue is going to be the fidelity between your virtual model and the experimental subject. Can your marker set be systematically translated into an OpenSim model accurately and reliably (and hopefully programmatically)?

You will need to decide what types of questions you expect the model to be able to answer, and whether you can do this with the marker set you plan to use.

David

Thank you David!

That's the reason why I have some concerns in using our protocol.
We use clusters and anatomical calibration to reconstruct the position of the anatomical landmarks that are most affected by soft tissue artifact as the asis and psis. The results is that we do not have these markers on the subject during the acquisition. We instead obtain a .trc file with the "corrected" landmarks thanks to a post-processing in Matlab.
I think in this case it may be harder to place the virtual markers on the model.
In fact, unfortunately, we do not have a code able to accurately and reliably reproduce the marker set on the model, taking into account the slightly differences at each acquisition session.

Thank you again.

Alice

Three years too late, but check this out:
http://biomechanical.asmedigitalcollect ... id=2554135