When performing a functional task (single leg bridge) until fatigue (non-isometric), can the surface EMG recorded signal be analyse using Entropy to determine the individual muscle resistence to fatigue during the task?
Sounds like an interesting possibility. Another possible approach would be to compute or estimate the fractal dimension of the surface EMG signal as a function of time. If one of the effects of fatigue is a decrease in signal variability (in the extreme it would correspond to the emergence of physiological tremor) or, equivalently, dimensionality, then one might expect the surface signal to show a reduction in dimensionality as fatigue developed with an effective loss of degrees of freedom and a reduced signal complexity. I was involved in doing some of this research some time ago in a 'short report' paper published in the journal 'Muscle and Nerve'. There may have been others who have looked at this since then.