Detection of Abnormalities in a Human Gait Using Smart Shoes

Abstract

Health monitoring systems require a means for detecting and quantifying abnormalities from measured signals. In this paper, a new method for detecting abnormalities in a human gait is proposed for an improved gait monitoring system for patients with walking problems. In the previous work, we introduced a fuzzy logic algorithm for detecting phases in a human gait based on four foot pressure sensors for each of the right and left foot. The fuzzy logic algorithm detects the gait phases smoothly and continuously, and retains all information obtained from sensors. In this paper, a higher level algorithm for detecting abnormalities in the gait phases obtained from the fuzzy logic is discussed. In the proposed algorithm, two major abnormalities are detected 1) when the sensors measure improper foot pressure patterns, and 2) when the human does not follow a natural sequence of gait phases. For mathematical realization of the algorithm, the gait phases are dealt with by a vector analysis method. The proposed detection algorithm is verified by experiments on abnormal gaits as well as normal gaits. The experiment makes use of the Smart Shoes that embeds four bladders filled with air, the pressure changes in which are detected by pressure transducers.