Fuzzy ROI Based 2-D/3-D Registration for Kinetic Analysis after Anterior Cruciate Ligament Reconstruction

Abstract

Rupture of anterior cruciate ligament (ACL) is a serious problem for playing sports, which causes in functional stability of the knee joint. To restore this problem, various operation techniques of ACL reconstruction are proposed. Thus, it is important to numerically characterize the knee kinematics after ACL reconstruction. Then, we proposed an analysis method to estimate the three-dimensional (3-D) knee kinematics. However, the estimation accuracy was not enough. Because the target image did not have high contrast, for example, at the boundary between the femoral bone and the tibial bone. Then, born regions can not be extracted preciously because the target image has low contrast. In this paper, we propose a fuzzy ROI (region of interests) based image registration. This method attend the region where has clear contour of bone region and ignore the region where has murky contour of bone region, by using fuzzy degree map which is assigned by the fuzzy region of interests (ROI).