An Automatic Method Framework for Personalized Knee Prosthetic Modeling Based on Kinematic Geometry

The shape of a knee prosthesis has an important impact on the effect of total knee arthroplasty. Comparing to a standard common prosthesis, the personalized prosthesis has inherent advantages. However, how to construct a personalized knee prosthesis has not been studied deeply. In this paper, we present an automatic method framework of modeling personalized knee prostheses based on shape statistics and kinematic geometry. Firstly, the average healthy knee model is established through an unsupervised process. Secondly, the sTEA (Surgical Transecpicondylar Axis) is calculated, and the average healthy knee model is resized according to it. Thirdly, the resized model is used to simulate the knee's motion in a healthy state. Fourthly, according to the target patient's condition, an excising operation is simulated on both patient's knee model and the resized model to generate an initial knee prosthesis model. Finally, the initial prosthesis model is adjusted according to the simulated motion results. The average maximum error between the resized healthy knee model and the patient's own knee model is less than 2 mm, and the average maximum error between the motion simulation results and actual motion results is less than 3 mm. This framework can generate personalized knee prosthesis models according to the patient's different conditions, which makes up for the deficiencies of standard common prostheses.