Development of a novel adaptive training tool that uses artificial intelligence to provide individualized biomechanical feedback for runners in the real-world.

Running is a primary exercise catalyst against obesity-related diseases, yet many runners fail to meet their fitness and performance goals due to overuse injury. Current popular mobile running apps and wearables track fitness metrics i.e. speed, heart rate and calories spent, but largely ignore safety aspects related to injury-risk i.e. asymmetry, instability, impact shock severity, or excessive loading. In this project we focus on advancing state-of-the-art by developing a novel adaptive training tool which harnesses empirically-based injury-risk metrics using body-worn accelerometers that are robust to outdoor running conditions. Additionally, we will employ user-augmented contextual awareness via real-time speech-to-text labeling to help steer machine intelligence towards personalized predictive models relating to individualized injury risk. This C3 projects extends from an ongoing inter-disciplinary C2 collaborative project between the Human Movement Biomechanics Research Group of KUL and the Subdivision Declarative Languages and Artificial Intelligence Group of KUL. If successful this adaptive training program will enable runners to make smarter training decisions and make a difference to thousands of runners hampered by reoccurring injuries.

Keywords:training

Disciplines:Orthopaedics, Human movement and sports sciences, Rehabilitation sciences