Questions Regarding Data Analysis

[Dylan Mayanja]

1.) When I am recording a video of a participant walking or squatting, their skeleton on the checkerboard is not moving in correspondence. Instead, the skeleton is still and doesn't move until the video recording is complete and uploaded. Is there a way to fix this? This also leads to another problem I've dealt with - I have run into an issue where I'll record a video but won't find out till after that the participant was not synched, and the trial failed (won't know until the end because the skeleton is still)

2.) The outcomes my team and I want to capture for our study are stride lengths, walk pace, and dynamic postural control. Could you please notify us which Y axis points correspond to these outcomes (example image shown in attachments)

3.) Sometimes, when I go through the process of recording the neutral pose and then proceed to the next step. The skeleton image is not placed on the checkerboard, instead it is is placed outside of the checkerboard but is still visible. Is there a particular reason for this, could this be a result of the angle at which the iPads are in (facing the checkerboard)?

4.) After downloading the data, I have had trouble accessing the the kinematics and geometry data within the OpenSimsData folder because I don't have the right application. What is the necessary application needed in order to access the kinematics and geometry data for the recordings collected?

[Antoine Falisse]

Hey,

1) It is not a livestream, that is the intended behavior. And yes, you will only find out whether everything went fine when processing is done. It is a sequential process. You record the videos, they are being processed, and then you can visualize the results. What you see on the screen corresponds to the trial selected in the menu on the left.

2) The outcomes you can plot are joint angles, so not what you are after. Most of the gait metrics you want can be automatically computed offline using our scripts: https://github.com/stanfordnmbl/opencap ... nalysis.py. In about one month, this will be possible in the web application too.

3) The checkerboard is just a representation of the floor. It is positioned relative to where you placed the actual checkerboard during the camera calibration. It is recommended to perform your activities close to where you positioned the checherboard during camera calibration. This should also lead to have the skeleton on the floor for visualization. But it is okay if it is not, it is just visual.

4) There is a README file with some information. The files are under OpenSim format. But you can open the files in excel too. Or use our scripts for data analysis: https://github.com/stanfordnmbl/opencap-processing/

Best,
Antoine

[Dylan Mayanja]

Thank you for the response! I have a few more questions regarding your response to my second questions.

Is it guaranteed to be available in a month? Also, how long is the data stored?

[Antoine Falisse]

It is not guaranteed, we are doing our best and that's our estimate. You can use the offline scripts already. Data is stored for as long as you do not delete them. Best, Antoine

[Dylan Mayanja]

Hello OpenCap team, I recorded a few walking videos and noticed that the video recording automatically stops at the 1 minute mark. For the study my team and I want to conduct, we want the participants to walk for 6 minutes. Is there any way where we can increase the duration of the recording video? Thank you!!

[Antoine Falisse]

Hello, sorry this is not possible. This would results in huge files that would take forever to process.
Do you need the kinematics for 6 minutes in a row? Please consider breaking down your 6-min trial in shorter sub trials.

Best,
Antoine