A wavelet multiscale based method to separate the high and low frequency components of mechanomyographic signals

The study of mechanomyographic (MMG) signals during dynamic contraction requires a criterion to select the cut-off frequency of the filter utilized to separate the low frequency (L.F) component (basically due to gross movement of the muscle or of the body) and the high frequency (HF) component (related with the vibration of the muscle fibers during contraction). To date, there is not an established criterion to carry out this selection. In this study, we propose a wavelet multiscale based method to aid to select a suitable cut-off frequency to separate correctly the LF and HF components. This method has been tested in an animal model, with the signal acquired during spontaneous ventilations with a capacitive accelerometer applied on the costal wall. This signal, as the MMG signals during dynamic contractions, has a LF component that is related with the movement of the thoracic cage, and a HF component that could be related with the vibration of diaphragm muscle fibers during contraction. The results obtained in the two respiratory tests analyzed indicate that cut-off frequencies around 10 and 3 Hz, respectively, must be employed to eliminate the LF component. The proposed wavelet multiscale method appears to be suitable to carry out a preliminary study of the MMG frequency content in dynamic contraction protocols.