Weekly Update in lieu of Recurring Meeting of Cleveland Clinic Core Team
Date: April 25, 2017 (Update for two weeks)
Provided by: Snehal Chokhandre
Ongoing Action Items:
- Ahmet
- Review drafts of data manuscripts for submission.
- Snehal
- Complete studies for tissue testing reproducibility.
- Continue tissue testing of Open Knee(s) specimens.
- All
- Continue segmentation of Open Knee(s) specimens.
Notes:
- Tissue testing.
- Snehal conducted repeatability tests for cartilage unconfined compression to assess effects of multiple freeze-thaw cycles. The cartilage strip, from which the sample was taken, was frozen and thawed 4 times already (including the specimen dissection). 6 tests were conducted over 6 days. Sample was frozen after every test. For the first test, 10g force position was used as the starting location for preconditioning and stress relaxation and therefore not included in the assessment. For the following 5 tests, preconditioning and stress relaxation tests were started 300 microns above the 10g 'find contact' location to capture the entire span where the force started to rise. This was done to aid with finding the actual zero force-displacement point.
- The first 3 tests in the set of 5 tests were repeatable with ~8% variability. The last two tests however were very different. The stiffness appeared to decrease significantly. The sample seemed to deal with multiple freeze-thaw cycles (8) before the properties changed dramatically.
- For the last two tests, the sample started to experience force at indenter positions earlier than that of the previous tests, i.e. the sample appeared to be stiffer at lower strains.
- The sample thickness variation was not different. Especially for tests 4 and 5, it was the same. Yet the tissue response was quite different.
- As the position where the sample started to experience force changes, the total strain applied was different for each test. If position for 10g load is used for application of strains then it remains consistent.
- For the first test in this set, Snehal accidentally entered incorrect starting position for the first stress relaxation ramp (instead of adding 300 microns to the 5% strain value (~100 microns), she added 200 microns hence the indentor was still 100 microns away from the 10g find contact position when the test started), hence the total applied strains were lower. Instead of discarding the data, Snehal continued the test. The only difference was that a maximum of 11-12% strain was applied from the 10g position instead of 15%. This will be rectified for any future tests.
- Since 10g was significantly lower than the force experienced by the cartilage sample at 5% strain, Snehal believes it is still a suitable value for establishing full contact with the sample surface and set as a starting point.
- Snehal modified the tissue data analysis to add a function to the Python script to find the actual zero force displacement. She is currently working on making the whole script modular.
- Snehal has also complied a literature review for knee cartilage testing and updated on the Specifications/ExperiementationTissueMechanics wiki page.
- Using PBS, longer wait times between tests, modified preconditioning protocol, and starting the stress relaxation tests before the contact cover all the measures to test the cartilage unconfined compression sample in a way to assess repeatability and provide various strategies for analyses.
- Snehal is currently conducting tests with the updated protocol with starting the preconditioning and stress relaxation 300 microns off the 10g contact position. Cartilage unconfined compression tests for the 48 yr old specimen over 4 days will be conducted with the sample kept in fridge between tests. Snehal plans to repeat this procedure for the 78 yr old test specimen. In the next 3 weeks, Snehal plans to conduct several sets of repeatability tests for all cartilage test specimen combinations.
- The unconfined compression test set for both the specimens (48 yr old and 78 yr old) will be followed by week long tests for cartilage confined compression and another week of tensile tests. All these tests will be conducted in PBS and over several days. Essentially all the modifications made so far will be covered.
- Once these sets are complete, Snehal and Ahmet will make a decision on whether to proceed with the actual Open Knee(s) tests or make further modifications to protocol.
- Data manuscripts.
- No progress.
- Segmentation.
- No progress.
- Other.
Snehal and Ahmet met last week to discuss the modeling & simulation proposal Snehal has put forth.
- Snehal would like to assess the effects of variations in image segmentation performed by different people and the resultant geometries on the finite element model predictions for patella cartilage, if any.
- Snehal proposed simulating variability in surface topology to represent variations in segmentation performed by different users on the same image set (using the same tools and protocols) and the effects on the peak contact stresses. Snehal proposed to obtain 6 geometries for the same sample from segmentation attempts by different people, averaging the surface and randomizing the master surface topology to simulate segmentation variations for a larger set.
- Ahmet was concerned that the error simulation (based on randomization) may not represent a real segmentation at all as user/ manual segmentation is guided by the visual of the image. He recommended using the actual segmentations by different people to compare the model predictions and then varying the material properties to further assess the variability in peak contact stresses.
- Snehal is currently working on building and running simple patella models.
- Snehal attended a two day training session for the VIC-3D strain measurement system. The system will not be used for the Open knee(s) tissue tests but the VIC-2D software can still be used to measure 2D strains from the images obtained by the Mach-1 system during tensile tests.
- Ahmet has been interacting with Cobus Muller. Cobus has started working with Open Knee(s) data set to develop models. He is from Stellenbosch University, South Africa visiting Trent Guess at University of Missouri. Cobus' initial efforts to work with Open Knee(s) kinematics-kinetics data has been informing Ahmet about navigation of the data.