Feasibility study of a smart insole based AI Gait Assistant system

Conference: BIBE 2019 - The Third International Conference on Biological Information and Biomedical Engineering
06/20/2019 - 06/22/2019 at Hangzhou, China

Proceedings: BIBE 2019

Pages: 5Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

An, Geng; Yang, Mingjing (College of Physics and Information Engineering, Fuzhou University, Fujian, China)
McCalmont, Graham; Morrow, Philip; Wang, Haiying; McClean, Sally; Zheng, Huiru (School of Computing, Ulster University, Jordanstown Campus, Northern Ireland)

Gait analysis is an important tool in healthcare it is used to obtain insight into the progress of several medical conditions such as Parkinson’s and osteoarthritis. However, gait analysis is usually based on the subjective opinion of a healthcare professional. Technical solutions have been developed to remove this subjectivity. One such system is the GAITRite mat, a grid of pressure sensors that can perform gait analysis. In this paper we propose a new low-cost insole-based AI gait assistant system that will be able to monitor patients for longer periods of time called AIGaitAssistant. In order to assess the performance of the AIGaitAssistant system a dataset was collected of 8 participants. The subjects were asked to walk on the GAITRite mat while wearing the smart insoles. A new gait detection algorithm is used to detect the number of steps, which is compared to the manually counted during the recording. The results from the GAITRite mat and AIGaitAssistant system were compared and the results show that the average values of the stand time and step time are the same and the maximum difference average value of swing time is around 0.07s indicating that there is no significant difference between the two systems in terms of major parameters calculations. This highlights the feasibility of developing a low-cost gait assistant system.