Highly Reliable Shear Sensitive Triboelectric Smart Textile for Intelligent Gait Analysis Toward Early Detection and Offloading Rehabilitation of Diabetic Foot Ulcers

Abstract

Diabetic foot ulcers (DFUs) are a severe complication of diabetes that leads to substantial morbidity and mortality. Quantifying compressive and shear stresses during the gait process provides an effective approach for assessing the early risk of DFUs and enabling timely intervention. However, current gait analysis devices suffer from a critical lack of shear stress detection capability, as well as limitations related to comfort, privacy concerns, and long-term stability. This paper proposes a shear stress sensitive triboelectric smart textile sensing system for intelligent gait analysis. The system comprises a smart insole featuring a triboelectric smart textile array in high-stress foot regions, a flexible printed circuit board (FPCB) for signal transmission, and a self-attention based multi-modal recognition software. An interlocked hill structure, inspired by mammalian skin, is integrated into the triboelectric smart textile. This structure helps focus and amplify localized stress, enabling rapid acquisition of external stimulus information with high sensitivity (0.062 V/kPa for shear stress, 0.05 V/kPa for compressive stress), fast response (less than 20 ms), stable (enduring 40000 cycles), and accurate sensing of a wide stress range (0-200 kPa). The compressive and shear stress cues obtained from the triboelectric smart textile are feature-extracted and model-fitted by a meticulously designed multi-label sequence data recognition neural network model based on self-attention mechanisms. Inter-module communication among multiple prediction heads enables deep gait-related insights, enhancing the models generalization, computational efficiency, and interpretability. The system supports patient identity recognition, early detection and offloading rehabilitation of DFUs, offering a low-cost, energy-efficient, and universally applicable solution for next-generation smart healthcare systems.