Estimation of recruitment and firing rate in EMG signals

We present a method for estimating the number of active motor units and the mean motor-unit firing rate for an intramuscularly recorded EMG signal. First, the motor-unit action potentials (MUAPs) are detected by high-pass filtering the signal and detecting all the spikes that exceed a threshold. Then the number of motor units, the mean number of spikes per MUAP, and the mean firing rate are computed using the autocorrelation of interspike intervals. Results from ten moderate-force EMG signals agree well with results obtained by computer-aided decomposition.