Page 2 of 2

Re: Extract Muscle Force Vectors (Lines of action)

Posted: Fri Mar 27, 2015 7:42 pm
by rosshm
Hi all,

I'm bumping this old thread because I have a related question.

I've previously used OpenSim to calculate muscle line-of-action (LOI) vectors using the method Ton described earlier in this thread (define translational coordinates at the joints, then ask OpenSim to calculate "moment arms" with respect to these coordinates; partial derivatives of muscle lengths with respect to the coordinates in question gives LOI vectors in the "child" segment's reference frame). This works great if your goal (for example) is to calculate hip joint contact forces and you need to know LOI vectors for all muscles that span the hip, in the femur reference frame.

What I'm trying to do right now is a bit different. I would like to calculate LOI vectors in the femur frame for all muscles that attach anywhere on the femur. I haven't been able to get the method described above to work for this. It misses out on the LOI vectors for muscles that originate on the femur (vasti, gastrocs) because their lengths are not affected by hip translation (shank and patella stay put because their parent is the femur).

The force directions plugin that Ayman mentioned earlier sounds like it would do what I'm looking for, but unfortunately it does not run in OpenSim 3.2.

Is there another way of accomplishing this within the <body> and <joint> definitions in a .osim file? The only other approach I could think of was to manually extract all the relevant muscle coordinates, wrapping definitions, and reference frames and do it in Matlab, which I was hoping to avoid =)

Ross

Re: Extract Muscle Force Vectors (Lines of action)

Posted: Wed Aug 12, 2015 6:34 am
by johnoconnor
Hi all,

This thread has been of great help so far. I'm trying to implement the method described above by Ton so i've added the translation DoFs to the hip in the model file and asked Opensim to return the moment arms for various muscles around the hip, I am getting unit vectors back but they all have a negative direction (i.e X,Y,Z components are all negative for all muscles) which doesn't seem to match what would be expected from looking at the model?

Can anyone help with this?

Thank you in advance
John

Re: Extract Muscle Force Vectors (Lines of action)

Posted: Fri Aug 14, 2015 6:27 am
by rosshm
Hi John,

Which model and what motion are you using?

Also which muscles specifically? All the hip muscles?

I would expect that maybe the z-directions could all be the same but it does sound odd that (x,y) would be the same as well for all muscles.

Ross

Re: Extract Muscle Force Vectors (Lines of action)

Posted: Fri Aug 14, 2015 6:59 am
by johnoconnor
Hi Ross,

Thanks for the reply, i'm using the model distributed with the Static Optimization example which is Gait2392 and the motion is subject01_walk1_Kinematics.sto

Seems to be all negative directions for all the hip muscles, the negative Y direction on the femur is proximal to distal in the model so this dosen't make sense for muscles spanning the hip.

Any suggestions would be very helpful

Thanks again
John