Using Wearable Soft Sensors for Gesture Recognition

Abdul-hussain, Gasak

Abstract

The increasing prevalence of hand impairments due to conditions such as arthritis, Cerebral Palsy, Parkinson’s Disease, and stroke presents significant challenges in everyday activities, such as tying shoes or getting dressed. In the UK, long-term musculoskeletal conditions are on the rise, highlighting the urgent need for effective rehabilitation methods. Despite physical therapy's potential to help regain motor skills, there is no consensus on optimal methods for promoting neuroplasticity. Robotic and wearable technologies have emerged as viable solutions, with soft robotics offering distinct advantages due to their flexibility, adaptability, and portability. However, limited evidence supports the superiority of conventional robotic devices over traditional therapies.

This PhD research investigates the development of a soft tactile sensor aimed at improving rehabilitation outcomes for individuals with upper limb impairments, focusing on muscle activity during hand movements in healthy, Parkinson’s, and stroke patients. The motivation for this study lies in the growing demand for accessible rehabilitation solutions that address the UK’s healthcare challenges, particularly for stroke survivors, where upper limb rehabilitation is under-resourced. The primary aim of this research is to design and validate a novel fabric-based tactile sensor using Eeon-Tex conductive stretchable elastic fibre, capable of accurately detecting muscle activity. The methodology includes the fabrication of the sensor, an investigation into the nonlinear hysteresis phenomenon, and validation against a commercial surface electromyography (sEMG) sensor. A key focus is on developing reliable alternatives to traditional sEMG systems, making rehabilitation more accessible.

Key findings demonstrate that the soft tactile sensor is effective in capturing distinct muscle activity patterns across various patients, particularly during tasks involving gripping and manipulating objects. Statistical analysis showed high signal similarity between the tactile sensor and sEMG, confirming the sensor’s reliability and potential for clinical application. Additionally, strategies were developed to mitigate the effects of nonlinear hysteresis on the sensor’s performance.

In conclusion, this research contributes to the field of rehabilitation technology by providing a cost-effective, reliable alternative to conventional muscle monitoring systems. The significance of this work lies in its potential to improve the quality of life for individuals with mobility impairments, particularly in an ageing population, while addressing the resource challenges faced by healthcare systems such as the National Health Service (NHS).

• Thesis Type: Thesis
• Deposit Date: Jan 13, 2025
• Publicly Available Date: Feb 24, 2025
• Award Date: Jan 23, 2025