Problem scaling ...what's the right segment?

[Caterina Passini]

Hi, I was trying to put some data I get from my lab in the model with only legs, I used the same data before for a complete model (just without arms) and the scaling process succeeded, now I was trying to do the same with just both legs, I cut off from files the marker I didn’t add to the model, but I’m not able to scale legs correctly (neither IK seems to work). I check the units of measurements, and every segments where I put the marker, so I have no idea why it doesn’t work. It seems like these legs have a different system of coordinates, but I tried also to change in my data the orientation of axes, but it didn’t still work…does anybody have some ideas? Thank you very much in advance!
Caterina

---I would like to add that I put the marker directly on my model, selecting the different segments, i.e calcn_R, tibia… but I’m a little bit confused since my data for static pose have coordinates referring to the same system of coordinates, so, if I use them in Opensim, is not possible having a good model scaled because Opensim works with different system of coordinates for every segment! I tried putting every marker just on the Calcn_l segments, and now the model looks perfectly scaled!(just that since the IK doesn't still work well, the legs are crouched) But what is the correct way to work? It doesn’t make a lot of sense put everything on the segment Calcn_L, moreover, with the previous model (Torso+legs) in which I used the same data I didn’t do that..I don’t know what to think…anyone does?

Thank you
Caterina

[Ayman Habib]

Hi Caterina,

Can you explain why you changed from one model to the other? Different models could have completely different number of segments, segment names, frames, marker names, etc. I don't think you can draw any conclusion about the models from just looking at them in the graphics window. The both legs model and other models used in tutorials (except for gait2354, 2392) are instructional only and may not have mass properties altogether which would prevent you from doing any dynamic simulation downstream, they also have only few muscles.

If you worked with one of the gait models but it has more segments than you want to account for, then you can lock coordinates or change mass properties to fit your needs without starting from scratch.

I'm not sure what you see on the scaleTool but some questions:
1. Are you using measurements or manual scales? if the earlier, you need to provide a static trial with marker names that match those in the model, and provide marker pairs used to scale individual segments.
2. Usually you do marker placement along with scaling to account for imprecise placement of markers on the subject, did you check this option?

Hope this helps and please let us know if you continue to have problems with scaling the model.

Best regards,
-Ayman

[Caterina Passini]

Hi Ayman, first of all thanks for your reply, actually I would like to use the both legs model since I noticed it has more muscles, so I think it better suit my exigencies. I have a .trc file with the marker that I put on the model(The same name), that contains the coordinates for each marker in time (I got them in my lab in a static trial).
I tried to use the tool as I did with the gait2354, so I gave legs a mass, I loaded my .trc file after I checked “Marker data for measurements” and “Adjust model marker”, and I think that’s all. The data I’m using I’m almost sure they are correct since are the same (they are a part) of which I used in the complete model. So I can’t understand what’s wrong…thank you very much
Caterina

________________

Oh, well the problem is solved, the model BothLegs can’t translate in the space, just rotate
Thank you

Caterina

[Steven Davidson]

Caterina,
Could you please explain what exactly you did to solve the problem? I have been running into the same issue and my adjustments aren't working.

Thanks,
Steve

[Ayman Habib]

Hi Steve,

When you solve IK (or do marker placement in scaling) you need to solve for the translations of the model to align with the experimental data. If the model doesn't have a degree of freedom to translate in space (as is the case with the bothlegs model) then the solution will fail. The gait model has degrees of freedom to translate the base segment (pelvis) relative to ground so it doesn't have this problem.

Hope this explains,
-Ayman

[Steven Davidson]

Got it, Thanks!

[Mark Lo]

Hi. I am experiencing the same problem with my model as well. It seems that my model lack the x,y,z axis displacement to shift the model origin to aline with the experiemntal data. Is there any way to enable this reposition?