Recurring Meeting of Cleveland Clinic Core Team

Date: April 14, 2015

Time: 10:30 AM EST

Means: In person meeting


  1. Ahmet Erdemir
  2. Craig Bennetts
  3. Snehal Chokhandre
  4. Robb Colbrunn
  5. Tara Bonner


  1. Discuss immediate action items from the last meeting.
  2. Discuss Mach-1 tissue testing system issues.
  3. Decide immediate action items for the next meeting.
  4. Other.

Immediate Action Items:

Ongoing Action Items:


  1. Discuss immediate action items from the last meeting.
    • There are many ongoing activities ongoing that do not necessarily fit into the immediate action items category. These items will be listed in the Ongoing Action Items category to ensure consistent (yet not time sensitive) effort on these.
    • Snehal contacted the imaging facility to schedule sessions for upcoming specimens. June 2, 2015 (Tue) experimentation was scheduled for June 3, 2015. The remaining dates for experimentation were not changed.
    • Work on the pivot shift manuscript is not urgent. Nonetheless, work on this activity should be conducted on a routine basis. This activity is assigned as an ongoing action item for Snehal.
    • Snehal and Craig conducted additional tests to compare probe-based thickness measurements to camera-based ones. Foam, cartilage, and tendon were measured. For optical measurements, the recordings were collected both on the back and front faces. As expected, the probe-based thickness measurements were lower. Ahmet proposed the group to build a simple model to assess potential deformation during probe-based thickness measurement. Snehal will follow up with this.
    • Ahmet created a wiki page for convergence optimization of Open Knee(s) models as part of the use cases. The goal is to increase simulation speed without compromising simulation results. The study will target at identifying the role of boundary conditions, tolerance criteria and other solver settings, material properties, and new model features on convergence. The initial approach will use passive flexion as the loading case and conduct a brute force grid search on simulation parameters. Craig will be working on a Python script to allow user to specify specify multiple parameters and multiple values for each parameter, e.g. 5 parameters, 5 levels => 5^5 simulations. When using one node with 16 threads, the group can conduct at most 18 simulations in parallel at the high performance computing facilities of the Cleveland Clinic.

    • Ahmet evaluated SimpleITK Python library to convert formatting of magnetic resonance images. He showed and described the Python code he wrote for this purpose. He will utilize the NIfTI format for dissemination of anatomical imaging, which also provides the data in a de-identified manner.
    • Ahmet and Tara met to acquire anatomical landmarks from Open Knee(s) - Generation 1 model. Tara provided the new pivot shift kinetic load curves, transformed from anatomical tibia to femur coordinate system.
  2. Discuss Mach-1 tissue testing system issues.
    • Both Snehal and Ahmet conducted tests with the Mach-1 tissue testing system. It appeared that the system has difficulties to apply desired high speed ramp displacements for the given displacement magnitude. Default acceleration setting for Mach-1 is twenty times the prescribed velocity, which constrains the lower bound for the desired displacement. The prescribed velocity also appears to be instantaneous (at mid time point of movement) rather than average. The paraboloid movement function in Mach-1 allows individually specifying acceleration and velocity but not the others. The paraboloid movement can be used for load and unload type of experimentation; starting the experiment before contact and move beyond target strain to extract data from constant velocity time range. The group decided upon asking Biomomentum, Inc. to see if acceleration can be set for different commands. Snehal will communicate this issue with them. If resolved, this capacity may also narrow down the time for deceleration, at which many specimens exhibited relaxation.
    • Another issue is related to the setting of controller parameters to ensure the testing system follows desired commands appropriately. While there is a way to do this, it is not interactive. Snehal will reach out to Biomomentum, Inc. to see if the interface to change controller parameters can be updated to evaluate the response immediately.
    • Ahmet described an alternative strategy to obtain tissue parameters. Rather than conducting a multi-step stress relaxation test, loading-unloading to a prescribed strain level at different rates can be accomplished. This may be possible with the paraboloid function; starting the experiment before contact and move beyond target strain to extract data from constant velocity time range. If necessary, these data may be used to reconstruct stress-relaxation at a desired strain level. The group will conduct trials to evaluate this type of experimentation and write up relevant specifications.
  3. Decide immediate action items for the next meeting.
    • See Immediate Action Items above.
  4. Other.
    • Ahmet informed the group that a high school student will be joining the project. This will result in a total of four students during the summer. Ahmet formulated the projects in more depth:
      • Python scripting to extract tibiofemoral joint mechanics data to be used for modeling & analysis; in related reproducibility of joint level experimentation for each specimen based on anterior-posterior laxity

      • Patellofemoral joint contact pressure analysis, as a function of quadriceps force and tibiofemoral flexion angle.
      • Optical measurement of tissue thickness (potential project for the high school student)
      • Registration to determine transformation matrices of rigid bodies between anatomical imaging and joint experimentation coordinate systems.
    • Additional group projects were also described:
      • Assisting convergence analysis studies, particularly for cases involving new model features.
      • Image segmentation to reconstruct tissue geometries, including segmentation from the same specimen to assess intra-/inter-observer variations.

RecurringMeetings/2015-04-14 (last edited 2016-05-04 22:09:52 by localhost)