Structure-Aware Fatigue Modeling in Foot Deformities: A Digital Health Framework for Tissue-Specific Risk Prediction Using Multi-Modal Data

Foot deformities are common across the general population, and hallux valgus is one of the most widespread forms. Although often perceived as a mild structural problem, hallux valgus can quietly change how people load their feet during daily activities. Over time, these altered loading patterns may accumulate and increase fatigue in key tissues—such as the Achilles tendon and plantar fascia—raising the risk of chronic pain and overuse injuries. However, evaluating these fatigue processes typically requires laboratory equipment and expert analysis, making early detection difficult and limiting widespread screening. To address this gap, we developed a digital health framework that uses wearable sensor data together with structure-aware fatigue modeling to estimate tissue-specific fatigue risk in people with foot deformities. This framework combines foot morphology, joint mechanics, and cumulative tissue loading to provide a personalized fatigue profile for each individual. We further built a hierarchical machine-learning model that predicts fatigue trajectories with high accuracy using only wearable data, enabling simple, non-invasive, and continuous monitoring outside specialized labs. Our approach offers an accessible method for identifying early signs of tissue overload and guiding timely behavioral or clinical interventions. By providing individualized, real-time fatigue assessment, this digital health framework has the potential to improve long-term foot health, reduce preventable injuries, and support large-scale screening in community and clinical settings.