High Hip Joint Forces

Hi,

I am trying to simulate a model of a liftig motion with the MoCapModel of the AMMR 1.6 (using AMS 6). Unfortunately the calculated joint forces in the hip are too high (>800% BW). I would be very glad if sb. could have a look at this model.

Thanks!

Hi Patrick,

the problem is your butt!

In that deep knee bend position gluteus muscles have a very bad moment arm and produce huge loads. I have seen that before and tried to implement a solution with muscle wrapping in the butt. please use the attached Mus.any file with slight adjustments for the gluteus wrapping and add following define to the body parts setup file:

// Wrapping of the gluteus maximus muscle can be added for both sides
// by including this define statement:
#define GLUTEUS_MAX_WRAPPING_BETA 1

this will reduce the loads to 4-5times BW. This is still a lot, but much more realistic. PLease be aware that this gluteus wrapping is “beta”, that means it has not been tested much in other circumstances.

Hi Amir,

thanks for you reply! That sounds really nice but unfortunately I receive some errors I don’t know how to handle atm. For the Mus.any AnyBody is searching for the definition of “TLEM_MORPHOLOGY” (#define-statement) which isn’t defined anywhere in my model. The “GLUTEUS_MAX_WRAPPING_BETA” on the other hand is not defined in the Mus.any so I don’t know this definitions purpose.

Regards,

Patrick

It should work in AMMR v1.6.

otherwise define it as 1

I’m sorry, I placed the file in the wrong directory, I just noticed it works fine if I replace the original Jnt.any in LegTLEM. The error only occured because I included it from MoCapModel/Model.

Another error occurs:
ERROR(SCR.PRS9) : E:/0_P…te/0…e/A…y/VP02/Body/A…n/L…M/Mus.any(2245) : ‘cyl’ : Unresolved object
There doesn’t seem to be a
“AnySurface &srf = …Seg.Pelvis.Muscles.GluteusMaximusWrapSurf.cyl;”…

updated file

Amir you’re a lifesaver, awesome! Works perfectly and gives much more realistic hip joint forces!

Hi Amir, Hi Patrick,

I just thought to give a hint/advice here as well: the diameter of the wrapping surface of the gluteus maximus has a significant influence on the muscle activation/forces and then of course on the hip reactions as well.

My question is: based on which data do you choose the value for the diameter?

Thanks,
Best regards,
Alex

Alex,

as mentioned above, it is beta and needs more testing. It is based on cadaver measurements from TLEM.

Unfortunately I have a problem with the hip forces again. Maybe you could have a look at this model? It’s a movement where the subject picks up a weight to it’s left and then puts it down to it’s right and I receive hip force up to 12xBW (about 9x in the attached trial). The forces for a lifting motion on the other hand are not to high for this model.

Patrick,

do you add the weight of the box somewhere? I couldn’t see it…

Hi Amir,

thanks for the reply. You didn’t miss anything, the weight isn’t included in the simulation. Of course I could just use the lower body model in this case but I want to have the option to turn the weight on using the same kinematics.

Patrick,

I don’t have a proper fix for this. If the subject is really carrying 25kg then it might make problems. If you try to lift 25kg yourself, you will experience trouble finding the balance. If you get out of the perfect position, you will not be able to lift the load. It should be the same for the model. Every in accuracy will increase the error.

  1. did the subject wear shoes? He is slightly above ground.
  2. Marker position on pelvis are quite important (pelvis tilt). You could try vary that.
  3. What happens when including the weight of the box?
  4. What happens when you analyze the same subject without a box?
  5. How convinced are you from your C3Ds? I know several GaitLab people try to run calibrations in AnyBody. C3D with some markers, specific weight on forceplates and see what AMS says? I guess you tried something for validation before?

Sorry to not have a better answer for now.

Amir, thanks for the answer.

The subject is actually me :wink: And lifting of 25kg, 40kg and 50kg was possible without touching the legs with the weight or losing balance, but of course the markers on the model or the segment length might be sligtly different which results in these problem.

  1. You are right, I was wearing shoes.
  2. I’ll have a look into that!
  3. & 4. No difference at all. Maybe if I move the hook somewhere at the head?
  4. The C3Ds should be OK, specific weights on the forceplates are well measured (at least in y direction) and the distance between two markers on a fixed staff is nearly constant and correct in the measurement volume. Also other motions ins the same lab of the same subject don’t produce such high forces.

I’ll go with #2 and try around a little.

I’ve got the impression, that the thorax doesn’t bend like it should. Is it possible that some dofs in the spine are restricted? I didn’t find anything like that but it seems like the pelvis was in a wrong position because the upper body cannot move like it should.

Patrick,

the Lumbar spine is moved by a spinal rythm, also the cervical spine, thoracic spine is currently only one fixed segment. Most of that is working fine for Gait, but might cause trouble for your extreme motion. Are you using the detailed neck? That could help a bit.

Amir,

do you have any more suggestions what I can do about this model?

Patrick,

unfortunately, I’m not aware of any easy fixes for that problem.