Scaling 3DGaitModel2354 without upper body markers

[Kate Mockler]

Hi all,

I was just wondering if it is possible to scale the generic 3DGaitModel2354 without upper body markers?

I have tried scaling it but it becomes bent over- is this because of my limited lower body markers? If yes, do I need to pick another model or is there a way of scaling this one so that it is upright? For now I have just loaded my markers and static .trc file- I have not adjusted scale factors from 1.

Thank you in advance!

Kate

[jimmy d]

The weird position of the trunk has nothing to do with trunk scaling and everything to do with marker labeling and IK posing. Your trunk angle, relative to the pelvis, has not been changed. The pelvis orientation is completely incorrect-- you will need to fix which markers you are using to pose the Pelvis. The same could be said for the knees. It is hard to give you a specific marker, or set of markers, to look at since it looks like the whole thing is incorrect. This could be caused by incorrectly labeled experimental markers, model markers, or both. Also could be due to bad tracking weights, or no weights at all.

[Kate Mockler]

Hi James,

Thank you for your response. You were right- my first attempt at marker placement was completely wrong. I had used a sacral wand (coda motion equipment) to collect data but when scaling I had placed virtual markers on the PSIS and ASIS which did not correspond to the sacral wand. I have since corrected.

Just a quick question in regards to inverse dynamics if you don't mind: My inverse kinematic .trc files and ground reaction force .mot files exported from the same .c3d file are different time durations: The .mot file is longer and the ground reaction force occurs around ~7 seconds, whereas the .trc file is only 6 seconds long. Am I right in assuming that this will affect the inverse dynamics as the kinematic gait cycle stops just before the force plate is reached?

When I preview the experimental ground reaction force .mot file, I can see the arrow clearly when it runs. But obviously when I try to associate the motion data, the gait cycle stops before it can be coordinated.

I assume I have to ensure that both files are the exact same time duration. If you could please confirm this for me that would be great.

[jimmy d]

The start and end time stamps don't need to be the same, but they need to have corresponding, synced, times for the portion of the trial that you wish to perform ID on. If you continue to have issues, please start a new thread and submit the trc and mot files in a zip.

[Kate Mockler]

I have corrected this and both files are now synced (used Matlab c3dexport this time). Force arrow is now displaying at correct time point.

Thank you for your help!

Kate