Contact and take-off

[Edilson Borba]

Hello,

I have some doubts about the processing carried out through the collections made in opencap.

When we need to determine the contact and take-off moments we use the data extracted from opencap and processed by the joint reaction in opensim. These patellofemoral joint reaction data return data that, when processed, we were able to automatically determine these moments. We performed a visual analysis to determine whether the method is reliable and is very sensitive to moments.

However, we are having problems explaining why this happens and why these values only work for running, and we would like to know if there is another way to perform this contact and take-off determination, both for running and walking, and if using these joint reaction data can really be an alternative.

If you can help us it would be a big step for our studies. If you need more details on how the calculations are performed, I can provide you with no problem.

Thank you very much

[Antoine Falisse]

Hey,

Thanks for your question. Could you elaborate on how you run the joint reaction analysis? Do you have force data?

Antoine

[Edilson Borba]

Hello,

We ran the analysis in the opens of the joint reaction, and extracted the patellofemoral data, below the data in the excel and the graph of the reactions. We understand that the negative values are the air phase, and the positive values of contact. When we perform the visual analysis the values are confirmed. However, we do not do any other analysis, and perhaps there may be an easier or more validated path than the one we used





Thanks

[Scott Uhlrich]

Hi,

I recommend using marker or kinematic features to segment gait cycles. We recently pushed a tool to do this for walking in opencap-processing. An example for walking is here https://github.com/stanfordnmbl/opencap ... nalysis.py, and you can see how we segment gait cycles here: https://github.com/stanfordnmbl/opencap ... is.py#L324. This segmentation approach (peaks and troughs in anterior-posterior position of heel/toe markers relative to pelvis) may work well for running as well, but we have not tested this.

[Edilson Borba]

Hi, thanks for information.

In the code (gait_analysis.py) there is an error in the use of the np.int function. As of NumPy 1.20, the np.int function has been deprecated and you must use the int type. It would be nice to make this change in the code so that others don't have the same error

Thank you very much for providing and helping with your posts and creations.

[Scott Uhlrich]

updated it. Thanks!