Improved semi-automated 3D kinematic measurement of total knee arthroplasty using X-ray fluoroscopic images

Abstract

Accurate measurement of 3D dynamic kinematics of total knee arthroplasty (TKA) is very important for evaluating the outcome of surgical procedures and for improving the implants design and clinical outcome. In order to achieve 3D kinematic analysis of TKA, 2D/3D registration techniques, which use X-ray fluoroscopic images and computer aided design (CAD) model of the knee implants, have been applied to clinical cases. However, in previous study, these techniques have needed time-consuming and labor-intensive manual operations in some process, and particularly the process of the manual initial guess poses of the CAD model for each X-ray image was problem for practical clinical applications. In this study, we propose an improved method for semiautomated 3D kinematic measurement of TKA using X-ray fluoroscopic images. The proposed initial pose estimation method of the model is based on the use of a transformation with feature points extracted from each X-ray image using speeded up robust features algorithm. In order to ensure the validity of the proposed method, in vivo experiment using Xray fluoroscopic images of 4 TKA patients during knee motions was performed. As a result of experiment, 3D pose of the model for all X-ray images except for the first frame was automatically stably-estimated, and the success rate (X-ray frame number within error of 1mm / 1 degree relative to all X-ray frame number) for the femoral and tibial component were 83.7 % and 73.5 %, respectively. In addition, the success rate with the proposed method was higher than a conventional method. Consequently, the present method was thought to be very helpful for practical clinical applications.