CMC error with different model

saylee patkar · Post by **saylee patkar** » Mon Mar 04, 2019 2:36 am

hi all,
I am trying to analyse the muscle forces. I first executed the analysis(scaling-IK-CMC) using gait2354 model but then noticed that the muscle vastus lateralis is not included in that model so I tried executing the analysis using same data files with gait2392 model. But using this model gives me an error as shown in the attached file below.
Why this error has occurred?
Also what is the significance of filter kinematics in CMC? we get different outputs with and without filter. So which output should be considered as feasible?
Note: I cannot share the data files due to confidentiality issue.

Dimitar Stanev · Post by **Dimitar Stanev** » Tue Mar 05, 2019 12:41 am

But using this model gives me an error as shown in the attached file below.
Why this error has occurred?

The error message alone does not provide any useful information. You can try to execute the analysis up to the point of failure and visualize the results to see whether the movement is tracked well enough.

https://simtk-confluence.stanford.edu:8 ... d+with+CMC (troubleshooting)

Also what is the significance of filter kinematics in CMC?

It is very important to filter the kinematics because the first and second derivative are calculated numerically. Otherwise, you will get spikes in the derivatives due to noise artifacts in the signal.

saylee patkar · Post by **saylee patkar** » Tue Mar 05, 2019 12:54 am

Hello sir,
I executed the simulation without RRA because after RRA the execution of CMC stopped at initial frame only.
Also, after scaling the Foot of the model is twisted and IK runs with the distorted foot. But this distortion increases after RRA.

Is it necessary to run RRA before CMC, since in the user's guide it is mentioned that to check the results we should check the kinematics results of RRA and CMC.

Dimitar Stanev · Post by **Dimitar Stanev** » Tue Mar 05, 2019 1:50 am

Also, after scaling the Foot of the model is twisted and IK runs with the distorted foot.

If you have kinematics issues then they will propagate in the dynamics analysis and cause problems.

Is it necessary to run RRA before CMC, since in the user's guide it is mentioned that to check the results we should check the kinematics results of RRA and CMC.

RRA is an optional (may be skipped) method that tries to reduce inconsistencies between measurements and model predictions.

saylee patkar · Post by **saylee patkar** » Tue Mar 05, 2019 1:57 am

RRA is an optional (may be skipped) method that tries to reduce inconsistencies between measurements and model.

Then, how to check the feasibility of the outputs? Since there is difference between vicon output and Opensim output. (I found in the forum posts that vicon doesn't have joints constraints and Opensim does therefore there is error between two outputs.)

Dimitar Stanev · Post by **Dimitar Stanev** » Tue Mar 05, 2019 7:12 am

between vicon output

What do you mean by vicon output? The marker trajectories?

https://simtk-confluence.stanford.edu:8 ... d+with+RRA

saylee patkar · Post by **saylee patkar** » Tue Mar 05, 2019 7:24 am

The vicon output for kinematic data for trial like knee joint angle, Hip joint angle, etc.

Dimitar Stanev · Post by **Dimitar Stanev** » Tue Mar 05, 2019 8:41 am

This comparison requires that both the Vicon and OpenSim joint definitions are the same. You cannot compare them unless they use the same model. But, even if they use the same model, IK may lead to different conclusions by changing the tracking weights in the optimization.

https://simtk-confluence.stanford.edu:8 ... tics+Works

CMC error with different model

CMC error with different model

Re: CMC error with different model

Re: CMC error with different model

Re: CMC error with different model

Re: CMC error with different model

Re: CMC error with different model

Re: CMC error with different model

Re: CMC error with different model