Abstract
Plantar fasciitis severely impairs daily life through persistent pain and limited mobility, whereas conventional treatments often lack real-time monitoring and personalized feedback. This study introduces a fully self-powered digital wearable system (FS-DWS), integrating an arch support auxiliary (ASA) device, a wearable sensing system (WSS), and a machine learning-driven closed-loop visualized feedback system (VFS) to enable real-time plantar pressure monitoring and abnormal gait recognition for the auxiliary treatment of plantar fasciitis. As a system-level engineering achievement, the ASA module integrates elastic support with energy harvesting, alleviating plantar pressure and powering the wearable sensing system without any batteries, with a maximum power density of 41.6 mW/cm(3), one order of magnitude higher than those of previously reported biomechanical energy harvesting devices. The VFS utilizes a flexible sensor array to collect dynamic pressure data, which is processed via a machine learning algorithm to achieve real-time classification of 7 gait cycle phases with an accuracy of 99.3%, enabling identification of abnormal pressure distribution, causal tracing of gait deviations, and generation of personalized correction instructions. As a proof-of-concept study, the proposed dual-function strategy of "physical support + intelligent regulation" provides an efficient and sustainable approach to the long-term management of plantar fasciitis and supports a shift in therapeutic approach from passive relief to active correction.