Crack-Expansion Patterned Laser-Induced Graphene Strain Sensors for Machine Learning-Assisted Neck Posture Monitoring

Prolonged exposure to electronic devices exacerbates neck health issues in modern populations. To address this challenge, we develop a flexible strain sensor based on crack-extension patterned laser-induced graphene (PLIG), designed for high-precision neck posture recognition. Laser-Induced Graphene (LIG), fabricated on a polyimide substrate, was subsequently transferred to a stretchable PDMS substrate, forming the piezoresistive strain sensor with high sensitivity. The successful transfer of Laser-Induced Graphene (LIG) from a polyimide film to a flexible polydimethylsiloxane (PDMS) substrate via a transfer printing method enabled the fabrication of the highly sensitive piezoresistive strain sensor. Through optimized laser processing and patterned structural design, the sensor achieves exceptional performance including a maximum gauge factor of 204.90, rapid response time (113 ms), and robust cycling stability (>5000 cycles). Integrated into an intelligent monitoring system, multichannel LIG signals processed via machine learning algorithms enable recognition of common neck postures with 99.45% accuracy. Further applications in human motion tracking and wireless manipulator control demonstrate the sensor's versatility in wearable electronics, human-machine interfaces, and telemedicine. This work delivers a scalable technical strategy for personalized health management and ergonomic intervention.