Time-frequency and principal-component methods for the analysis of EMGs recorded during a mildly fatiguing exercise on a cycle ergometer.

Electromyographic signals contain the information on muscle activity and have to be frequently averaged, compared, classified or details need to be extracted. A time-frequency analysis, based on wavelets, was previously presented. The analysis transformed an EMG signal into an EMG-intensity-pattern showing the intensities at any point in time for the frequencies filtered out by the wavelets. The purpose of the present study was:to define and apply a new EMG-pattern-space for the analysis of EMG-intensity-patterns; and to determine the variation of EMG-intensity-patterns while getting mildly fatigued by cycling on a cycle-ergometer. The coordinates spanning the pattern space were principal components of the measured EMG-intensity-patterns. A point in pattern-space thus represented an EMG-intensity-pattern. Fatigue resulted in points moving along a line in pattern space. The line was characterized by an intercept at time 0 and a slope. Thus mild fatigue caused a shift from an initial intensity-pattern representing the intercept to a final intensity-pattern adding gradually larger amounts of the pattern representing the slope. The intensity-pattern of the slope revealed the physiologically important individual strategies for coping with mild fatigue. Changes were observed at different times and at different frequencies during the cycling movement.