Recurring Meeting of Cleveland Clinic Core Team

Date: October 21, 2014

Time: 10:30 AM EST

Means: In person meeting


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


  1. Discuss immediate action items from the last meeting.
  2. Discuss coordinate systems and transformation matrix information required to build and run FE models.
  3. Discuss oks002 testing plan.
  4. Decide immediate action items for the next meeting.
  5. Other.

Immediate Action Items:


  1. Discuss immediate action items from the last meeting.
    • Progress on tissue testing was provided.
      • Snehal updated tissue testing specifications. In particular, she described how the loads are applied in a way to collect data at different frequencies during hold periods of stress-relaxation. She also organized specifications for each type of tissue. Robb and Tara will read and review these specifications for next week.
      • Snehal was not able to finish meniscus testing for oks004 as she was waiting for the O-rings to prevent leakage of the tissue bath. O-rings are received and the tissue testing can continue.
      • For oks004 tissue components, Snehal changed the sample labeling to conform data management strategies.
      • The group emphasized the need to consolidate joint and tissue testing tools (grips, supplies) potentially by listing them in a database. There are issues with various tools, fixtures, parts missing. Also some parts may return as damaged, i.e., some of the screws stripped tapped holes of the tension grips. To prevent this, all were advised to use torque wrench when using tension grips. Various ways to keep track of supplies were discussed, e.g., labeling, logging, sign-out sheet. Robb and Tara will provide a sign-out sheet to exchange of tools between laboratories.
    • Updates on oks001 experimental data were discussed.
      • Snehal provided all oks001 data on in-house Midas platform for organization.
      • Tara provided laxity curves for oks001. The data seem to be in order and the laxity response appears to be as expected.
      • Snehal filled out various sections of the oks001 wiki page. There are not many issue and deviations for joint level imaging and testing.
      • All group members were asked to fill in information on the parts they were in charge of (including notes, deviations, team members): Robb and Tara - tibiofemoral joint testing (include laxity curves), Robb and Ahmet - patellofemoral joint testing, Snehal - specimen preparation and dissection, Snehal and Craig - imaging
    • For summarizing coordinate systems, Snehal and Craig will need another meeting with the group (also see the following agenda item).
  2. Discuss coordinate systems and transformation matrix information required to build and run FE models.
    • Ahmet started the discussions by telling the desired inputs to the model.
    • All the models will likely be in the image coordinate system. From imaging data, one can establish registration marker coordinate systems for each bone, e.g. femur.
    • Data from robotics joint testing provides transformation matrices between the registration marker coordinate system of a bone, the optotrak coordinate system, and the anatomical coordinate system. The joint coordinate system of robotics testing, e.g. for the tibiofemoral joint, includes the anatomical coordinate system on tibia, an optimized anatomical coordinate system on femur, and a convention to describe joint movement in between. When the transformation matrix between registration marker coordinate system of a bone and its anatomical coordinate system is known; the same anatomical coordinate system can be placed in the image (therefore in the bone). In this manner, one can describe movement axis on each bone using the location of these axis relative to the registration markers, e.g. a flexion axis on femur, an internal/external rotation axis on tibia, and a varus/valgus axis in between.
    • Once movement/loading axes are defined in the model (exactly in a way as in the experiment), kinematic/kinetics measurements of robotics testing can directly be applied to the model to prescribe tibiofemoral joint loading and deformation. For example one can apply tibiofemoral joint movements to the axes, simulations can predict joint reaction loads, which can later be compared to experimental data. Alternatively, one can apply tibial forces and moments, simulations can predict joint translations and rotations that can be directly compared to clinical movements reported by the experimentation. One needs to accommodate the offsets in kinematics (rotations and translations). First, the reported movements are referenced to the neutral state of the joint in the robot. Second, the reference state of the model is as imaged and will likely be different than that of the experiment. Therefore one needs to calculate absolute state of the joint configuration for the model's reference state and for the experiment's reference state, and during movements of experimentation to appropriately prescribe kinematics in silico.
    • Snehal wondered how the transformation matrices between a rigid body's Optotrak, registration marker, and anatomical coordinate systems were obtained from raw data. During specimen preparation both the pose and orientation of a rigid body's Optotrak coordinate system and the position of a point of interest (anatomical landmark or registration marker) are acquired in the stationary global Optotrak coordinate system. Using the transformation matrix between moving and global (World) Optotrak coordinate systems, one can describe the point in the local Optotrak coordinate system of the rigid body. Using three points acquired in a similar fashion, one can establish a local registration marker or anatomical coordinate system (and the transformations in between). Robb noted that all these digitized points are provided in global (World) coordinate system of the Optotrak paired with origin and orientation (yaw, pitch, roll) of the local (bone) Optotrak coordinate system.
    • All digitized points a in the physical world coordinate system; all transformation matrices are in the right-knee abstraction convention. Robb will add details in the documentation about transformation matrices in the configuration files.
  3. Discuss oks002 testing plan.
    • The group agreed that oks001 testing went smoothly. Nonetheless, there were a few minor issues that need to be accommodated for oks002 testing.
    • Snehal and Ahmet will take specimen out on November 2, 2014, this time earlier in the morning, e.g. 9-10 AM.
    • For specimen preparation (scheduled for November 3, 2014) at least three people are needed. It is also advised to start earlier in the morning, e.g. 9 AM, as long as the specimen is thawed.
    • oks002 specimen is smaller and lighter, which may be easier to tackle when compared to oks001.
    • The logistics of bringing a surgeon for specimen dissection need to be taken care of. Specimen dissection is scheduled for November 5, 2014. Morgan Jones and his residents/fellows have a hard time to accommodate Wednesday afternoons. Jack Andrish or a research resident may be able to help.
    • Leaders for each task will be assigned next week.
  4. Decide immediate action items for the next meeting.
    • See Immediate Action Items above.
  5. Other.
    • None noted.

RecurringMeetings/2014-10-21 (last edited 2016-05-04 22:09:52 by localhost)