A muscular fatigue index using superimposed M waves and fuzzy rules

A practical muscle fatigue index is proposed in this paper based on the correlation between the instantaneous frequencies (IFs) of the superimposed M wave and the mean power frequency (MPF) of the preceding background activity. The authors investigate the details of the distribution of a feature vector (MPF, IF) in the two-dimensional space. The experimental results showed that the correlation coefficients between MPF and IFs was high during a first phase of a sustained contraction and them they became uncorrelated as muscular fatigue progressed. Combining the correlation coefficients and conventional fatigue indices, the authors propose a fuzzy rule based muscular fatigue index. The fuzzy rule based muscular fatigue index demonstrated an exhausted stage clearer than conventional indices, whereas conventional indices saturated