Dual-Network PVA/PAM Hydrogel Strain Sensor for Machine-Learning-Assisted Rehabilitation-Oriented Hand Motion Monitoring.

Wearable rehabilitation monitoring requires soft strain sensors with mechanical robustness, stable electromechanical responses, and intelligent motion recognition capability. Here, we report a poly(vinyl alcohol)/polyacrylamide (PVA/PAM) double-network hydrogel strain sensor for rehabilitation-oriented wearable monitoring. The hydrogel was prepared by ultraviolet ray (UV)-initiated acrylamide polymerization followed by freeze-thaw-induced PVA crystallization, forming a covalent PAM network interpenetrated with a physically crosslinked PVA network. The resulting hydrogel possessed a compact porous structure, improved stretchability, and stable deformation recovery. The optimized sensor exhibited a tensile strength of approximately 0.52 MPa, an elongation at break of approximately 480%, a response time of 0.12 s, and a recovery time of 0.17 s. It generated repeatable resistance signals under cyclic strain, finger bending, wrist motion, and grip training. Furthermore, the sensor enabled morse-code information transmission and support vector machine (SVM)-based recognition of rehabilitation-related hand states, including straight, bend, and clench. This work provides a soft hydrogel sensing platform for real-time rehabilitation-oriented hand motion, while morse-code encoding provides auxiliary assistance and an emergency communication function.