A facilely prepared notch-insensitive nanocomposite organohydrogel-based flexible wearable device for long-term outdoor human motion monitoring and recognition. Issue 6 (26th January 2023)

• Record Type: Journal Article
• Authors: He, Zhejian; Hua, Haoqiang; Zou, Zhantao; Shu, Lin; Wang, Tao; Sun, Weixiang; Xu, Xiangmin; Tong, Zhen
• Is Part Of: Journal of materials chemistry. Volume 11:Issue 6(2023)
• Page Start: 2316
• Page End: 2327
• Publication Date: 2023-01-26
• Subjects: Materials -- Periodicals; Chemistry, Analytic -- Periodicals; Optical materials -- Research -- Periodicals; Electronics -- Materials -- Research -- Periodicals; 543.0284
• DOI: 10.1039/d2tc05038b
• Languages: English
• ISSNs: 2050-7526

Abstract

Abstract : A notch-insensitive and environment tolerant PAA-clay-glycerol organohydrogel strain sensor was facilely prepared and assembled to obtain an integrated wearable device for long-term human motion monitoring and classification. Abstract : Flexible wearable hydrogel devices show great potential for applications in human health monitoring. However, facile preparation procedures, reliability when defects are present, and the realization of actual wearable devices are still severe challenges. This work presents a facile method to prepare a reliable nanocomposite organohydrogel (NCO hydrogel) strain sensor and the corresponding flexible wearable device for long-term monitoring of outdoor human motion. The NCO hydrogel was facilely prepared through the in situ free radical polymerization of acrylic acid in a clay suspension, followed by immersion of the hydrogel in glycerol. The hydrogel shows excellent notch insensitivity, universal adhesion and environmental tolerance. More importantly, the hydrogel strain sensor with different notch lengths can still output steady electrical signals similar to that of the unnotched sensor, which shows favorable reliability required in actual applications. We designed a portable wearable device by attaching the NCO hydrogel to the substrate and connecting it to a Bluetooth module for the long-term outdoor monitoring of human motion. Various human physical motions can be recorded by the device and real-time displayed through an APP on a mobile phone. A two-level fully connected neural network (FCNN) was used to perform the classification and recognition of different motions through deep learning with a classification accuracy of 87%. The accuracy does not decrease even with a notch on the hydrogel sensor. We have provided a new approach to prepare actual reliable wearable devices through functional hydrogels for long-term outdoor human motion monitoring and recognition in the human health field.