An EMG system for studying motor control strategies and fatigue

We present a system for polymyographic analysis which addresses detection of muscle fatigue and strategies assumed by the central nervous system to deal with it. The system consists of EMG amplifiers, force transducers, A/D converter, portable computer and software running in the LabView environment that allows real-time and detailed offline processing of EMG signals in time and frequency domains. We demonstrate the features of the system by using the example of analyzing the strategy to generate 80% percent of the maximum force for prolonged period of time. Force sensor was used to detect muscle fatigue (fall of the force bellow the selected threshold), and EMG recordings were used for the analysis which of the quantitative measures of EMG is correlated with this. We tested the following four methods of EMG measures: 1) median frequency, 2) short-time mean frequency, 3) mean frequency of scalogram and 4) fractal dimension. We show that the system is capable of providing reproducible results and could be used for diagnostics and basic research in motor control. The analysis shows that the median frequency used often is not the best predictor of fatigues, and the measure needs to be selected based on the relative activity of the muscle compared to its maximal activity.