Abstract
Computer-assisted orthopedic surgical procedures routinely require the registration of a bone to preoperative surgical planning data. The registration procedure requires the collection of points on the bone that are then matched to points or surfaces on a 3-D bone model. To improve the registration algorithms, simulations of the real-world registration processer are performed to test the changes to the algorithms. The present publication proposes the use of generative adversarial networks (GANs) to generate registration simulation data that closely matches that of real-world experimental/actual data.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
THINK Surgical, Inc., "Generative Adversarial Networks (GANs) for Simulating Human Behaviors during Bone Registration to Improve Registration Algorithms", Technical Disclosure Commons, (June 05, 2020)
https://www.tdcommons.org/dpubs_series/3297