Effects of the physiological parameters on the signal-to-noise ratio of single myoelectric channel

BACKGROUND: An important measure of the performance of a myoelectric (ME) control system for powered artificial limbs is the signal-to-noise ratio (SNR) at the output of ME channel. However, few studies illustrated the neuron-muscular interactive effects on the SNR at ME control channel output. In order to obtain a comprehensive understanding on the relationship between the physiology of individual motor unit and the ME control performance, this study investigates the effects of physiological factors on the SNR of single ME channel by an analytical and simulation approach, where the SNR is defined as the ratio of the mean squared value estimation at the channel output and the variance of the estimation. METHODS: Mathematical models are formulated based on three fundamental elements: a motoneuron firing mechanism, motor unit action potential (MUAP) module, and signal processor. Myoelectric signals of a motor unit are synthesized with different physiological parameters, and the corresponding SNR of single ME channel is numerically calculated. Effects of physiological multi factors on the SNR are investigated, including properties of the motoneuron, MUAP waveform, recruitment order, and firing pattern, etc. RESULTS: The results of the mathematical model, supported by simulation, indicate that the SNR of a single ME channel is associated with the voluntary contraction level. We showed that a model-based approach can provide insight into the key factors and bioprocess in ME control. The results of this modelling work can be potentially used in the improvement of ME control performance and for the training of amputees with powered prostheses. CONCLUSION: The SNR of single ME channel is a force, neuronal and muscular property dependent parameter. The theoretical model provides possible guidance to enhance the SNR of ME channel by controlling physiological variables or conscious contraction level.