Development of a mathematical model that predicts optimal muscle activation patterns by using brief trains

Because muscles must be repetitively activated during functional electrical stimulation, it is desirable to identify the stimulation pattern that prod uces the most force. Previous experimental work has shown that the optimal pattern contains an initial high-frequency burst of pulses (i.e., an initia l doublet or triplet) followed by a low, constant-frequency portion. Patter n optimization is particularly challenging, because a muscle's contractile characteristics and, therefore, the optimal pattern change under different physiological conditions and are different for each person. This work descr ibes the continued development and testing of a mathematical model that pre dicts isometric forces from fresh and fatigued muscles in response to brief trains of electrical pulses. By use of this model and an optimization algo rithm, stimulation patterns that produced maximum forces from each subject were identified.