Recurring Meeting of Cleveland Clinic - University of Utah

Date: September 24, 2014

Time: 2:00 PM EST

Means: Conference call

Attendees:

1. Ahmet Erdemir (Cleveland Clinic)
2. Craig Bennetts (Cleveland Clinic)
3. Jeff Weiss (University of Utah)
4. Ben Ellis (University of Utah)
5. Steve Maas (University of Utah)

Agenda:

1. Update on feature implementations.

   1. In situ strain - http://wiki.simtk.org/openknee/Febio/InSituStrain

   2. Set definitions - http://wiki.simtk.org/openknee/Febio/SetDefinitions

   3. Connector elements - http://wiki.simtk.org/openknee/Febio/RigidBodyMotions

2. Upgrade of Open Knee(s) - Generation 1 to FEBio v2.0.
3. FEBio issues during simulations.
   1. FEBio v2; segmentation fault during solution.
   2. FEBio v2.1; incompatibility with rigid body BCs.
4. Decide action items for next meeting.
5. Other.

Immediate Action Items:

• Cleveland Clinic (Ahmet & Craig)

  • Prepare a test problem to evaluate set functionality.
  • Provide a sample problem to University of Utah team which exhibits segmentation fault issue.
• University of Utah (Steve)
  • Evaluate general purpose formulation of the in situ strain feature.
  • Setup a meeting with Ahmet and Gerard at CMBBE 2014 to discuss rigid body dynamics related issues, features, etc.

Notes:

1. Update on feature implementations.
   • Steve completed the general purpose implementation of the in situ strain feature. In situ strains can be prescribed as an initial multidirectional deformation field or as fiber stretch; both for coupled and uncoupled constitutive model formulations. He is currently running test problems, e.g. transversely isotropic ligament, to resolve convergence issues. Steve will present this feature in the upcoming Computer Methods in Biomechanics and Biomedical Engineering (CMBBE) meeting. Jeff received from Steve a draft of the methods section for a scholarly article; a manuscript submission is anticipated within the next few months. A proof of concept case study may be introduced in to this article using Open Knee(s) model, e.g. ligaments with pre-tension.
   • Evaluation of the set definitions feature is pending. The Cleveland Clinic team still needs to develop and run a simple test problem.
   • Rigid body dynamics related features will be discussed in the upcoming CMBBE conference in October, 2014. Steve will arrange a meeting with Ahmet and Gerard at that time for these discussions.
2. Upgrade of Open Knee(s) - Generation 1 to FEBio v2.0.

   • Cleveland Clinic team successfully upgraded the Open Knee(s) - Generation 1 model using PreView. A few warnings were issued by PreView during import of the model with the older version. Craig reported these warning to Steve through e-mail. In addition, the output requests in the model with the older version were not transferred to the upgraded model during conversion. Nonetheless, the simulations ran without further manual changes in the model file. Upgraded model and the description of the conversion process are available in the source code repository under Generation 1 related source folder.

3. FEBio issues during simulations.
   • Segmentation fault may be related to the version of the Intel MKL libraries, particularly for FEBio compiled from source code. Jeff advised the use of the binaries distributed at the FEBio site. Running simulations using single thread (with Pardiso) or by switching the linear solver to Skyline may help identify the source of the problem. Screening memory usage during simulations will be helpful. Cleveland Clinic group will send a test problem to Steve so that he can evaluate if it runs on their systems.
   • Ahmet and Steve will discuss the incompatibility of rigid body boundary conditions in FEBio version 2.1 with Gerard, at CMBBE conference in October, 2014.
4. Decide action items for next meeting.
   • See Immediate Action Items above.
5. Other.
   • None noted.