Hi,
We are beginning a study looking at OpenCap in children. Do you envisage any issues with this? I trialled it on a 9 year old yesterday who was 133cm tall and I could not get a static pose calibrated. I have set the cameras up again this morning exactly as they were yesterday (same room, location, lighting and network) and I have been able to collect data on myself without any issues.
Is there something you can recommend I look at from a calibration perspective? Or is the size of the child going to be an issue? We intend on collecting data on typically developing children and children with Cerebral Palsy.
Kind regards,
Denni
OpenCap in a paediatric population
-
Scott Uhlrich
- Posts: 124
- Joined: Tue Jan 20, 2015 4:29 pm
Re: OpenCap in a paediatric population
Hi Denni,
First off, I recommend using the HRNet pose estimation model (can select in Advanced Settings at bottom of page for static pose). This should help. The participant needs to stand straight and still for the calibration to go through, ideally ~3m from cameras. We have used the tool successfully in populations aged 13-18, but haven't rigorously tested in younger populations.
If the children with CP are using assistive devices, the pose estimation models will not do as well. Additionally, if they are unable to straighten their legs for the neutral trial, the scaled model may not be great but could be corrected by changing the opensim static pose setup file in post processing.
Feel free to apply to the Restore Center virtual office hours to discuss your project in greater detail.
Thanks,
Scott
First off, I recommend using the HRNet pose estimation model (can select in Advanced Settings at bottom of page for static pose). This should help. The participant needs to stand straight and still for the calibration to go through, ideally ~3m from cameras. We have used the tool successfully in populations aged 13-18, but haven't rigorously tested in younger populations.
If the children with CP are using assistive devices, the pose estimation models will not do as well. Additionally, if they are unable to straighten their legs for the neutral trial, the scaled model may not be great but could be corrected by changing the opensim static pose setup file in post processing.
Feel free to apply to the Restore Center virtual office hours to discuss your project in greater detail.
Thanks,
Scott