Muscle overload during inverse kinetic analysis

#1

Hi
My C3D model shows muscle overload during inverse dynamics analysis. Could you please tell me how to solve this problem? Thanks a lot for your help。

Best wishes
zzs

#2

Hi @zzs22500

It is hard to say exactly what is causing this. Can you check if the magnitude of force from the force plate makes sense?

Also, an a thing I immediately notice is that the pelvis appears too wide. This most likely because the marker position on you model doesn’t match the marker positions in your MoCap experiment. The system then tries to minimize the marker error by making the pelvis too wide.

BW
Morten

Error when loading C3D model
#3

Hi Morten
Thank you for your reply. What should I do to match the marker position on my model to the marker position in my MoCap experiment? And I have this warning when loading the model: Some of the detected transitions in the data set were skipped (i.e. not entered into ‘Ts’) because they too close compared to the transition time ‘dT’. 5 transitions were skipped! When I delete the LPSI marker, the warning disappears.

Best wishes
zzs

#4

There is no easy solution to that. But if you did the MoCap experiment your self, you may be able to judge if the marker postion in the model is identical to the virtual positions in the model. If not you need to move the model markers. Markers like ASIS/PSIS are supposed to be on the bony landmark. But the reality is that they are often quite far away. Often because of adipose tissue on the subject or because it is just difficult to place the markers correctly in the experiment.

That is because the marker has drop-outs. The AnyMoCap framework handles that by reducing the weight of the marker to zero before the marker drops out, and then fading the weight up again after the marker comes back. However, when the marker drops in/out several time very quickly there isn’t enough time to transition the marker weights, and those fast transition will skipped. It basically groups several drop outs as one long drop-out. That is not a problem, and you can probably ignore that warning.

BW
Morten

1 Like
#5

Hi Morten
Thank you very much for your reply.I have tried to match the virtual position with the marker position.I don’t know if it’s better to make these two markers closer.This is my model and C3D file.I deleted some markers because there are some markers drop out.Could you please tell me where is the problem?

BW
zzs

#6

Hello,

I usually try to run the parameter Identification to check if the markers are generally at the right position. If it failed at the 0 step which means that there is a big distance between some markers. If it can run several steps, you can stop it to see Which markers are far from the model markers, That could save your time. You could change the initial position of legs to see whether the markers are at the right place. And the heel markers and patella (Knee) markers and RASI, RPSI, LASI, LPSI markes are the good markers to determine the positions.

And there are no markers on your upper model, that could also cause the failure of Parameter Identification.

Hope this helps.

BW,
Lee

#7

@zzs22500 Just a quick thought to your original question about overloaded muscles in the model.

I think it could also be caused by the Psoas muscles trying to provide the necessary force to the trunk and upper body. That is really not what we want here since we don’t have any markers on the upper-body. So the dynamics could be completely wrong.

What you should do is add reaction-forces to all the joints of the upper body. Only then can we disregard the upper body completely.

Try to add the following to your model:

  Main.HumanModel.BodyModel.Trunk ={
    AnyReacForce SpineJointReactions={
      AnyKinMeasure &jnt1=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.T1C7Jnt;
      AnyKinMeasure &jnt2=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C7C6Jnt;
      AnyKinMeasure &jnt3=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C6C5Jnt;
      AnyKinMeasure &jnt4=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C5C4Jnt;
      AnyKinMeasure &jnt5=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C4C3Jnt;
      AnyKinMeasure &jnt6=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C3C2Jnt;
      AnyKinMeasure &jnt7=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C2C1Jnt;
      AnyKinMeasure &jnt8=Main.HumanModel.BodyModel.Trunk.JointsCervicalSpine.C1C0Jnt;
      AnyKinMeasure &jnt9=Main.HumanModel.BodyModel.Trunk.JointsLumbar.L5SacrumJnt;
      AnyKinMeasure &jnt10=Main.HumanModel.BodyModel.Trunk.JointsLumbar.L4L5Jnt;
      AnyKinMeasure &jnt11=Main.HumanModel.BodyModel.Trunk.JointsLumbar.L3L4Jnt;
      AnyKinMeasure &jnt12=Main.HumanModel.BodyModel.Trunk.JointsLumbar.L2L3Jnt;
      AnyKinMeasure &jnt13=Main.HumanModel.BodyModel.Trunk.JointsLumbar.L1L2Jnt;
      AnyKinMeasure &jnt14=Main.HumanModel.BodyModel.Trunk.JointsLumbar.T12L1Jnt;
    };
  };

I think that will solve your problem :slight_smile:

Inverse analysis failed when used AnyMocap with my own C3D data
#8

Hi Morten
I just added these to the model.It solves this problem.The inverse dynamics was successfully completed.Thank you very much :grin:

BW
zzs

1 Like
#10

Hi Lee

 Thank you for your help.This helps me a lot:grin:

BW
zzs

#11

Hi Lee,
I noticed that you also work with Mocap model, I want to ask you some questions about this, I hope It would not bother you.

  1. Did you use all markers as included in Mocap templates?
  2. If you did not use all, did you comment the marker you did not include in your C3D file?
    I tried to comment all upper body markers as suggested, but I got one unknow error every time I did it.
    Do you have any idea about this?
    Waiting for your reply, thanks.
#12

Hello,

I do comment on the markers which are not included in my own C3D files.
Actually, I created some new markers. And it worked well.

When did you get the information, after reload or run parameteridentification?

BW,
Lee

#13

Thanks for reply. This error occur when I run parameteridentifications. I do not know why. But one staff from Anybody suggests me to delete all upper body markers because I failed parameter identification for a squat trial.