Ligaments with custom stiffness function

[Erica Montefiori]

Dear OpenSimers,

I recently started playing with ligaments in my ankle model and I have a few doubts/issues I hope you will help me with.

1. I tried two different implementations: the Ligament class, defined with pcsa_force, resting length and force_length_curve, and the CustomLigament class (https://simtk.org/projects/multikneedss/) defined with resting length, damping, el region and stiffness. In the first case, I am not sure I completely understood the concept of pcsa_force to scale the force_length_curve. Also I looked in some models available on SimTK and the force_length_curve of the ligaments seems to behave very different from what I would have expected from physiology literature. Therefore, I preferred to opt for the second implementation but my problem is that the stiffness here is a constant number, whereas I would like to define it as a function of the length. Here my first question: is there a way to introduce a spline to customise the stiffness? Alternatively, could I use a spring class for my purpose? I tried to briefly look into this but before digging I was wandering if anyone else has used this approach to model ligaments and if there are models available so that I am facilitated in this research and implementation. I must specify that unfortunately my knowledge of C language is very limited so an example would be very beneficial.

2. I run my batch simulations using the Matlab API and I realised that, although the ligaments are included within the Force Set in the .osim file, they are not visible when accessing the Force Set class from Matlab. Anyone knows why this is happening and how I should access them?

3. When running the SO from the GUI I can see the ligaments forces in the force .sto file together with the muscle forces. When using the API and running the same SO (same setup file) from batch the ligaments forces are not there anymore. Any suggestions here? Should I update my setup file with different settings if using the API with a ligament model? Is this related to the issue in the previous point at all?

4. It is not clear to me whether the unit of measurement of stiffness is N/mm or something else. I would say it is, but in some published models I have found that they set this parameter to values that look very different from literature reported values, and this made me doubt.

Thank you in advance to the patient readers and hopefully I will get some good advice!

Erica