Gait analysis by using Smart Socks system

Aegean International Textile and Advanced Engineering Conference 2018

Eizentals, P.^1 ; Katashev, A.^1 ; Oks, A.^2

Institute of Biomedical Engineering and Nanotechnologies, Riga Technical University, 1 Kalku Street, Riga

LV-1658, Latvia^1

Institute of Design and Technology, Riga Technical University, 1 Kalku Street, Riga

LV-1658, Latvia^2

关键词: Disease detection;    Gait pattern;    High dimensions;    Manhattan distance;    Plantar pressure measurement;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/459/1/012037/pdf
DOI  :  10.1088/1757-899X/459/1/012037

来源: IOP

【 摘 要 】

Feet plantar pressure measurement and gait pattern analysis can be utilized for different disease detection as well as post-operation rehabilitation monitoring. Increasingly popular for gait analysis in research have become smart sock systems as an alternative to the more expensive commercial analogs, however, analysis with such systems are complicated due to the limited sensor count and other inherit limitations of textile pressure sensors. This paper proposes a new method for gait pattern recognition for use with smart sock systems. According to this method, steps are classified depending on the sum of Manhattan distances (SMD) between two high-dimension vectors, from which one describes the step to analyze and the other is a predefined vector representing a specific gait pattern. The gait pattern with the lowest SMD value to the particular step vector is assigned to each step. An experiment was performed to verify the proposed method, where three volunteers simulated three gait types, normal, pronation and supination, and the simulated steps were compared to references. The experimental results confirmed that typically the lowest SMD value was obtained by the reference corresponding to the tested gait pattern type. In one of the demonstrated cases, the mean SMD value for the same set of steps with the reference for normal step was 1.26, 1.77, and 1.78, with reference for pronation it was 1.75, 1.57, and 2.59, and reference for supination, 1.75, 2.15 and 1.46, for normal gait, gait with pronation and gait with supination, respectively. It can be seen that the lowest mean SMD value for each gait type is when the correct reference is applied.