A new intelligent model for automated assessment of elder gait change

This paper addressed a novel intelligent model for automatic evaluation of the change of elder gait function based on kinematic gait data. In order to recognize the change of elderly gait patterns with higher accuracy, the wavelet analysis technique was proposed as a new approach to extract gait features, and then those obtained gait features were initiated the training set of gait classifier such as artifical neural network (ANN). The gait data of two groups including young and old subjects were acquired during normal walking, and were analyzed using the proposed method. The experimental results indicated that the gait features exacted by the wavelet analysis technique, as the input of ANN, could provide more discriminating information than the traditional gait features selected such as maximal value or values obtained from the different occurrences based on gait events, and the proposed classification model could identify young and elderly gait patterns with higher accuracy. It is hopeful that the proposed model can be used as an effective tool for diagnosing the change of gait function for old people in clinical application.