Page 1 of 1

Marker placement after scaling

Posted: Thu Apr 27, 2023 6:10 am
by karosae
Hello,

I am using OpenSim 4.4 on Windows and I am facing a few issues while scaling the model.
  • During scaling, the Static Pose Weights are set to be: markers on bony locations, i.e., medial humeral epicondyle, both femoral epicondyles, both malleoli have the weight of 20.0, tracking markers (on the thighs and shanks) are set to have the weight of 0.0, and the remaining markers have the weight of 1.0
  • The patient was severely overweight, therefore the hip markers could not be placed on bony landmarks of the pelvis and instead the patient wore a belt with 4 markers, which were placed in the approximate locations of both left and right anterior superior iliac spine and posterior superior iliac spine
Attached you can find a few screenshots of the scaled model in a static pose (frontal and sagittal planes), together with the associated motion data.

Why is the head of the model tilted forward? The virtual markers (set as free in the marker set-up file) are matching the experimental ones, but not any more their appropriate location on the head, e.g., the markers on the back of the head are not even on the models head after scaling. The error I get is for the left marker on the back of the head: max = 0.172174 (lb_hd). I feel like if the head was in the right position, the whole torso of the model would not be so much bent forward and I would have a better fit.

Furthermore, why are the shoulder markers (originally placed on each acromion) drifting down? Originally they are placed on top of the acromion.

What I have tried so far:
  • Setting the head and the shoulder markers fixed, as well as free
  • Not giving any weight for the head markers during scaling (in the Static Pose Weights tab)
  • Adjusting the location of the virtual markers
Any help is much appreciated.


Thanks a lot,
Karolina Saegner

Re: Marker placement after scaling

Posted: Tue May 02, 2023 10:21 pm
by kernalnet
Hi, some tips and videos can be found here: https://simtk-confluence.stanford.edu:8 ... leshooting