보행 지표 추정에서의 천골과 질량 중심 운동 정보 활용의 비교

Abstract

The center of mass (CoM) is used as an indicator for gait monitoring or to estimate the walking data. The kinematics of CoM have been derived through calculation or estimation, because the CoM is not a fixed point on the body. Therefore, since it is difficult to measure the kinematics of the CoM, researchers have used the sacrum to replace CoM for monitoring using wearable devices. In this study, we compared the estimation results for walking data using the sacrum and CoM to discuss whether the kinematics of sacrum were sufficient to replace the kinematics of CoM. Seventeen healthy subjects walked at four speeds on a split-belt treadmill. The kinematics of sacrum were measured by a motion capture system, and the kinematics of CoM were measured by force-plates. To compare the kinematics of the sacrum and CoM in the estimation, an artificial neural network model with a single hidden layer was used to estimate the three-dimensional ground reaction forces and the two-dimensional center of pressures. Input data were set as the three-dimensional position, velocity, acceleration, and time of sacrum and CoM, respectively, for the learning and verification. The CoM kinematics showed more accurate estimation results than the sacral kinematics. This indicates that more accurate calculation or estimation of CoM kinematics are required for gait monitoring instead of the sacral kinematics.