Leather-Based Shoe Soles for Real-Time Gait Recognition and Automatic Remote Assistance Using Machine Learning.
Peng Zhang1,2, Xiaomeng Zhang1,2, Ming Teng1,2
1National Demonstration Center for Experimental Light Chemistry Engineering Education, College of Bioresources Chemistry and Materials Engineering, Shaanxi University of Science and Technology, Xi'an 710021, P. R. China.
ACS Applied Materials & Interfaces
|November 1, 2024
Summary
This study introduces a novel flexible sensor for real-time gait monitoring. The smart shoe system accurately detects falls and recognizes various gaits, enhancing health monitoring and human-computer interaction.
Area of Science:
- Materials Science
- Biomedical Engineering
- Wearable Technology
Background:
- Real-time gait monitoring is vital for healthcare but faces challenges in data acquisition and analysis.
- Existing methods often lack the stability and sensitivity required for effective application.
Purpose of the Study:
- To develop a flexible sensor for accurate and stable real-time gait characteristic monitoring.
- To create an intelligent system for human-computer interaction and fall detection using gait analysis.
Main Methods:
- Fabrication of a carbon nanotube/graphene composite conductive leather (CGL) sensor on a collagen fiber substrate.
- Integration of the CGL sensor into smart sports shoes for collecting foot motion signals.
- Application of machine learning, including an optimized K-Nearest Time Series Classifier (KNTC), for gait analysis and fall detection.
Main Results:
- The CGL sensor exhibits a high dynamic range (0.6–14.5 kPa) and sensitivity (S = 0.2465 kPa⁻¹).
- The optimized KNTC algorithm achieved 99% accuracy in fall detection with a 13 ms prediction time.
- The system demonstrated 90% gait recognition accuracy across diverse populations with low error rates.
Conclusions:
- The developed CGL-based sensor and smart shoe system offer stable and accurate real-time gait recognition.
- This technology provides valuable insights for plantar behavior monitoring and contributes to advanced healthcare applications.
- The system enhances human-computer interaction and shows significant potential for patient rehabilitation and telemedicine.
