Dynamic analysis of human knee

Abstract

Knee joint weakness and disorders are extremely common; with development of a new methodology for early detection of knee disorders in mind, various dynamics tests have been conducted. Subjects were requested to perform exercises of certain patterns including squats and marching at a constant rate for 70 (s). A force plate, acoustic sensor, and set of electromyographic (EMG) sensors were used to measure the ground reaction force and torque to the foot, the vibration from the knee cavity, and the electrical activity of muscles. The acquired signals were analyzed in the frequency domain to associate with certain biomechanical parameters. Preliminary studies indicated clear correlations between several features of the Fourier spectra and biomechanical parameters. The force and torque signals from a knee with a torn meniscus showed considerable broadening in the Fourier spectra around the peak of the exercise frequency (e.g. the 1 Hz peak of the torque associated with the knee extension from 1 Hz marching) and higher transfer function to other degrees of freedom, indicating instability of the injured knee. The acoustic sensor signal from an arthritic knee indicated significant attenuation due to the fluid in the knee cavity, leading to the extremely poorly defined spectral feature. The femoral efficiency, defined by the transfer function from the EMG signal to the force/torque of the dominant degree-of-freedom, showed clear correlation to fatigue in the amplitude and frequency.