Abstract - Bursting neurons play an important role in motor pattern synchronization and neuronal communication resulting in the ability to coordinate muscles across large anatomical distances. Such synchronization among multiple motor units is revealed by rhythmic surface electromyographic (sEMG) activity along paraspinal muscles. The Discrete Wavelet Transform (DWT) has long been used to explain motor unit recruitment by means of choosing a mother wavelet similar to the motor unit action potentials (MUAPs). However, the periodicity of the DWT and variability in the return times of bursts may cause misalignments between multiple MUAPs and the mother wavelet during the spiking events. For this reason, we developed an expert system with nested conditional statements each within a for loop that compares multi-delayed DWT sub-signals with their sEMG signal in order to time-localize and characterize the spiking events by increasing their waveform matching. In the present case study, the entire spiking events are best reconstructed at level 8 from two relatively high detail coefficients that span ~60 ms each, suggesting a high incidence of double spikes within each burst with a prolonged intraburst period (exceptional doublets).
This paper was presented at 40th International Conference of the IEEE Engineering in Medicine and Biology Society Engineering in Medicine and Biology Conference 2018 (EMBC'18) - July 2018.