Deep-learning-enhanced metal-organic framework e-skin for health monitoring
Xinyi Ke, Yifan Duan, Yifei Duan, Zhe Zhao, Chunyu You, Tingting Sun, Xingyu Gao, Ziyu Zhang, Wen Xue, Xuanyong Liu, Yongfeng Mei, Gaoshan Huang, Junhao Chu
Abstract
Electronic skin (e-skin) mimics the sensing abilities of human skin and offers sensitivity and flexibility, which has garnered attention for medical health monitoring. Nonetheless, limitations in the properties of active materials and signal-to-noise ratio hinder the implementation of multi-functional detection such as integrating biomolecular sensing, motion detection, and electrocardiography in a single device. Here, we introduce a strategy of preparing a multi-functional e-skin that utilizes a composite material combining self-locking chitosan and conductive metal-organic framework film, achieving high-performance motion detection and biomolecular sensing. The integration of deep learning, specifically transformer neural networks, aids in recognizing subtle facial micro-expressions. A sensing array is crafted by utilizing a rapid assembly method that supports discerning more attributes with heightened precision.