A framework for modeling motion of total knee replacements

This dissertation presents a framework for the study of passive knee kinematics. Computational models for solving both the forward and the inverse kinematics of a prosthetic knee joint were formulated. For forward kinematics, the model computed the predicted knee motion as a function of joint parameters; these parameters included patient-specific ligament data as well as the design and the surgical placement of the prosthesis. The model also provided three-dimensional graphics visualization of the predicted kinematics. It was applied to two commercially available prosthetic designs and was validated experimentally. Simulation results suggested several heuristic rules for surgical strategies for maximizing the postoperative knee function.;For the study of inverse kinematics, the computational model provided a means to estimate the patient-specific ligament data from a sequence of knee motion. The novel approach of the inverse kinematics model was that it embedded the forward kinematics model in an unconventional numerical-optimization paradigm. Given the knee motion and an initial estimate of the ligament data, the inverse kinematics model recovered a set of joint parameters that, when used as inputs to the forward kinematics model, resulted in the observed knee motion. Simulation results suggested that it may be possible to obtain patient-specific ligament data non-invasively, thus allowing the surgeons to use postoperative knee function as a surgical consideration. Together, the forward and the inverse modeling techniques described in this dissertation serves as a foundation for kinematic-based total knee replacement surgery.