Effects of Anatomical Frame Definitions on Repeatability of Knee Joint Translation Measurement During Walking

Human body movement analysis using stereophotogrammetry has been widely applied in understanding the normal function and pathological conditions of human knee joint. To reconstruct the 3-dimensional (3D) position and orientation of bones from markers placed on body surface, the spatial relationship between anatomical frames and skin marker clusters needs to be identified [1]. Since the determination of anatomical landmark (AL) locations in general lacks accuracy and precision, anatomical frames defined based on ALs are subject to uncertainty. This is a major issue in human movement analysis and its effects on knee rotation kinematics determination have been studied [1]. However, few studies have investigated the effects of uncertainty of anatomical definition on translation measurement of knee join. Although knee translations are secondary motions with only small magnitudes, they are regarded with great clinical significance. On the other hand, translation measurement is more sensitive to definitions of origin positions in addition to axis directions of the anatomical frames. In this work, we studied the effects of three different anatomical frame definitions (neutral standing, prediction method and functional method) on knee joint translation measurement during level walking. The repeatability of translation measurement among three sessions was compared among the three anatomical definition methods.