Deep Neural-Assisted Flexible MXene-Ag Composite Strain Sensor with Crack Dual Conductive Network for Human Motion Sensing

TL;DR

This paper introduces a new strain sensor with high sensitivity and a deep learning system for accurately detecting human movements.

Contribution

A novel MXene-Ag composite strain sensor with a dual conductive network and deep learning integration for motion sensing.

Findings

The sensor achieves GFmax = 487.3 and a response time of ≈65 ms.

Sensor arrays identify joint movements with up to 95% accuracy using deep learning.

The design maintains biocompatibility and comfort for health monitoring.

Abstract

Developing stretchable strain sensors that combine both high sensitivity and a wide linear range is a critical requirement for health electronics, yet it remains challenging to meet the practical demands of daily health monitoring. This study proposes a novel heterogeneous surface strategy by in situ silver deposition on modified PDMS followed by MXene spray coating, constructing a multilevel microcrack strain sensor (MAP) using silver nanoparticles and MXene. This innovative multilevel heterogeneous microcrack structure forms a dual conductive network, which demonstrates excellent detection performance within GFmax = 487.3 and response time ≈65 ms across various deformation variables. And the seamless integration of the sensor arrays was designed and employed for the detection of human activities without sacrificing biocompatibility and comfort. Furthermore, by adopting advanced deep learning technology, these sensor arrays could identify different joint movements with an accuracy of up to 95%. These results provide a promising example for designing high-performance stretchable strain sensors and intelligent recognition systems.

Figures