The Assessment of Human Gait Function Based on Wavelet Entropy

WU Jian-ning; WU Bin; WANG Jue

Abstract：Objective To seek a useful index for assessing the change of human gait function based on the application of wavelet entropy in the gait data analysis. Methods The ground reaction force (GRF) gait data of 2 groups( 10 young and 10 elderly subjects) were acquired with Bertec 3 dimensional force measured system (a strain gauge force platform) during normal walking. The orthogonal discrete wavelet transform was used for decomposing the collected gait data. The relative wavelet energy and wavelet entropy were defined and calculated respectively. They were statistically analyzed with t - test technique respectively to evaluate the change of human gait function. Results There was significant difference between the relative wavelet energy from GRF gait data of the two subject groups. The average values of wavelet entropy from the GRF gait data of elderly subject group were obviously less than those from the GRF gait data of young subject group. With the value of significance level of t-test P < 0.05, there was significant difference between wavelet entropy of 2 groups.Conclusion The relative wavelet energy can provide the detailed information associated with the change of human gait function in gait data, and the wavelet entropy is able to characterize the degree of the intrinsic change of the human locomotive system. The wavelet entropy can serve as a new index to describe the change of human gait function in future clinical application.