Development and evaluation of a computer-supported home exercise coaching system with real-time feedback for individuals with chronic knee pain

Abstract:

Knee pain is a disabling condition affecting many people, particularly older individuals. The high prevalence and the growing challenge call for effective approaches to improve treatment and care for individuals with knee pain. Exercise therapy is significantly effective as a key management strategy. Nowadays, technology-supported exercise programs are emerging as promising tools for facilitating remote exercise training and monitoring. The exercise therapy is becoming more accessible, affordable, and attractive.

Following the current trend, this study will develop an interactive exercise coaching system and assess its effectiveness for individuals with chronic knee pain. First, a systematic review and meta-analysis will be conducted to examine the effects of technology-supported exercise programs. The review study will also provide evidence of the effects of specific features which can better facilitate and encourage users to perform exercises. Second, in order to provide real-time feedback as an important feature, the RGB sensor (e.g., webcam) will be applied as an accessible device to support the exercise coaching system. The application of state-of-the-art pose estimation algorithms and machine learning techniques will allow real-time correction of exercise movements. Together with other educational and motivational features, this system will promote physical and emotional engagement in exercise therapy. Third, a randomized controlled trial (RCT) will be conducted to investigate the effectiveness of such exercise coaching system for individuals with chronic knee pain.