Een Machine-Learning Benadering om Belasting en Loopgerelateerde Blessures te Voorspellen op basis van Data van Wearables

Running-related overuse injuries (RROIs) impose a substantial burden on the running population, and preventing their occurrence remains an important challenge. The overall aim of this PhD project is to advance the prevention of RROIs by developing load monitoring and injury prediction models suited for real-world running conditions. Load plays a crucial role in RROI development. However, traditional load estimation methods are confined to laboratory settings, preventing continuous load monitoring during running in real-world conditions. Therefore, this project estimates structure-specific load and ground reaction force characteristics during running based on inertial measurement unit (IMU) data using a machine learning (ML) approach. Additionally, this thesis investigates the potential to predict running-related overuse injuries, enabling timely intervention and potentially preventing their occurrence. Future RROI occurrence is predicted using ML models based on IMU data and person-specific characteristics.