Feature Extraction of Knee Joint Vibroarthrographic (VAG) Signal Using Statistical Means

The purpose of the present preliminary investigation is to characterize the vibration signal between normal and symptomatic knee joint. Accelerometers were attached to the medial and lateral retinaculum of dominant knee to acquire the vibration signal induced during flexion and extension of leg. Signals were compared in time and frequency domain and statistical analysis were performed using mean, variance, standard deviation, skewness and kurtosis. The results show that the frequency range for the symptomatic VAG signal is higher than the normal. The results confirm that the proposed approach is indeed feasible for the non-invasive diagnosis of knee pathology disorders.