Complexity-based analysis of the correlation between stride interval variability and muscle reaction at different walking speeds

Abstract

In this research, for the first time, we evaluated the correlation between the variations of leg muscle reaction and gait at different walking speeds. Since leg muscle reaction in the form of Electromyogram (EMG) signals and stride interval time series (as gait variability) have complex structures, we utilized fractal theory and sample entropy to decode their alterations at different walking speeds. Twenty-two subjects walked at three different speeds (slow, comfortable, and fast) in six trials, and we analyzed the fractal dimension and sample entropy of EMG signals and stride interval time series. Based on the results, increasing the walking speed causes lower complexity in EMG signals and stride interval time series. Besides, strong correlations were found among the changes in the complexity of EMG signals and stride interval time series at different walking speeds. This method can be applied to analyze the correlation between other complex physiological signals of humans (e.g., EEG and ECG) during walking and running.