Matching Kinematics for TLEM1 & TLEM2

#1

Hello dear colleagues,

Recently we started an investigation.
The aim is to compare the impact of the cadaver data (TLEM1 and TLEM2) on the simulation results. In particular, the influence of the muscle and ligament attachments.

We want to achieve the same kinematics for TLEM1 and TLEM2 by using the same joint center locations and a set of additional markers.

What we already know:

  1. We need an pre-process operation sequence in the AnyBody script, as it is used for the BVH motion capture format.

  2. This operation reads .BVH motion capture data into the Anybody Software

  3. Marker trajectories and landmarks are created for every time frame of the BVH data

  4. After this preprocessing the data is saved in an output file [BVH_Trajectories.anyset]

  5. Run analysis computes the joint and muscle forces using inverse dynamics

Our Approach:

  1. Read .C3D data into the AnyBody Software

  2. Run parameter identification with the TLEM2 model for optimization

  3. To achieve uniform kinematics: Use a custom code (based on the preprocessing like the one for .BVH files) and save marker trajectories and marker positions (of joint centers and additional markers) in an .anyset file [new]

  4. Create an operation (that is included into the run analysis study) which uses the .anyset file of the preprocessing if the user wants to analyze data with TLEM1, achieving the same kinematics as in the TLEM2 parameter identification. [new]

What we would like to know is:

  • How can we save the marker trajectories of the joint centers and additional markers as anyset file?

  • Which markers (additional to those of the joint centers) are necessary to achieve a stable inverse dynamics simulation?

  • Which parts of the BVH operation sequence in the preprocessing file do we actually need for our approach?

We hope that you can help us to answer these Questions and that the amount of Information is sufficient.

Thanks in Advance.

Lukas

#2

Hi Lukas

Could we maybe do an web meeting or something similar. I think that would be easier. Then maybe we can work out the best solution together.

When do you have time?

Best wishes,
Morten

1 Like