Further questions about scaling

[Katie Ewing]

I read the papers from Wu et al. that formed the basis for the ISB standards on defining joint coordinate systems, but I'm not sure if this is what you are talking about. I still don't quite understand how I can accurately place the markers on the generic model, which must be very important if some markers are fixed during marker placement (more on that below). For example, for the R ASIS and L ASIS, I'm just guessing the locations off the pelvis geometry in the generic model (see below picture of my marker set). The marker coordinates are some arbitrary value not based on actual calculations but rather where they look like they should be on the model, and they don't include the soft tissue that may be present. I feel like the knee markers are even more arbitrary. Is there some sort of guideline that provides a "rule of thumb" for the virtual marker locations in OpenSim?

Let's assume my generic model markers are correct. How do I account for differences in my experimental anatomical markers (assuming they are also accurately placed) due to soft tissue? For example, some subjects may have a bit of fat around their knees, which will push their experimental knee markers out relative to the virtual markers. Do I still want to fix the virtual markers in the model, even though I know they are not representative of the actual experimental locations? This is also the case for the pelvis of some subjects with more body fat. I just don't know how critical it is to fix certain markers when the goal is to match the virtual markers to the experimental ones. In these cases, even if I do have confidence in them, they will still be different from the virtual markers, which doesn't necessarily mean "something is up."

Forgive me for keep asking all these questions. Please let me know if they don't make sense. I am just really trying to figure out how to make scaling work across all kinds of participants!

Thanks very much!
Katie

[shuyang han]

Hi Kaewing,

I actually have the same question. Recently I created a self-use markerset by adding several markers on the Gait2354 model (two are the R.PSIS and L.PSIS). I just give them a random location (-0.06 0.12 0.13, and -0.06 0.12 -0.13). After scaling, IK calculation, I got some weird results. For example, while the subject is standing in T pose, the hip flexion is -25deg. Have you found any guideline on definition of new marker locations? Thank you!

Cheers,
shuyang