Researchers at the University of Florida are using software for prototyping mechanical systems to study stress and motion in natural and artificial knees. It lets them combine a range of imaging and computational techniques with traditional CAD tools and reverse-engineering software. They want to identify knee implant designs or surgical positioning issues that contribute to wear and failure in artificial knees.

Total knee replacement (TKR) is a common remedy for the painful and debilitating effects of osteoarthritis, a degeneration of the cartilage covering the ends of the tibia (shin bone) and femur (thigh bone). In a TKR, cartilage-bearing surfaces of the tibia and femur are replaced with artificial surfaces. Surgeons fit the femur with a metallic bearing surface and the tibia with a metallic base plate. Although TKR usually restores mobility, plastic-insert wear often limits the implant's life span to 20 years or less.

Knee simulator machines, usually useful research tools in wear studies, have several drawbacks. For instance, a single series of tests can cost tens of thousands of dollars and take months. Loads are difficult to determine and identical implants tested in the same machine often produce different results.

“The goal is to create detailed models of artificial and natural knees and study wear in these joints under real-life loading,” says B.J. Fregly, assistant professor of mechanical and aerospace engineering. He thinks a computational wear model will explain and accurately predict joint wear and failure. “We are developing computer models that will help us see which implant and surgical positioning work best for each patient.”

Fregly's team begins by video recording movement data. The motion-capture system gives a quantitative picture of overall movement, but its accuracy can be thrown off by skin and muscle movement.

To create models of natural knees, Fregly's team uses CT scans, which produce static 2D image slices from the top to the bottom of a patient's leg. CT data imported into image-processing software is graphed as a 3D point cloud by defining the edges in a stack of 2D images of the patient's leg. The problem then becomes turning the point cloud into an accurate 3D model of bone surfaces.

Fregly's team uses Geomagic Studio software from Raindrop Geomagic, RTP, N.C., (geomagic.com) to convert point-cloud data from a scanned knee into a detailed polygonal model. The software also lets Fregly create surfaces well suited for shape-related tasks such as image matching.

After developing contact-stress predictions from movement data, the final comprehensive wear model is created with help from Greg Sawyer, a friction and wear specialist at the university. Combining knee-motion data with contact stress predictions creates a wear model that pinpoints exact places where an artificial knee is likely to experience problems.

The team has been able to compare wear predicted by its computational approach with actual wear on a recovered artificial knee. Predictions came within 0.1 mm of the actual maximum wear depth and accurately pinpoint where the worst wear would be.