A Method for Non-contact Measurement of Knee Load in Sport Motion

In this paper, we propose a method for non-contact measurement of knee load. Joint position data is acquired with image analysis. Masses and centers of gravity on each segment of human body is estimated with a multiple linear regression model. Knee torque is estimated based on kinetics and these data, and it is used index value of knee load. In the proposing method, human body is not restricted and measurement may be performed in a ordinary space without any special equipment set on the floor or wall. The method is effective for preventing knee injuries. The validity of the proposing method is verified with cross-correlation function between the estimated knee load and EMG signals. The knee load increases along with the knee flexion and extension, and is sustained by the torque around the knee exerted by the quadriceps femoris. EMG signal measured on the quadriceps femoris is also index value of knee load. The experiments are performed based on those data. As a result, it is reported that the time derivative of the torque corresponds to the EMG signals in the sense of phase.