With wearable health sensors rapidly developing, respiration is gaining increased attention as an essential human health parameter. Alongside respiratory changes occurring throughout life and exercise, symptoms like sneezing, nasal congestion, and coughing may indicate asthma and lung diseases. They also represent crucial indicators for early disease diagnosis and human health assessment.

To meet the increasing demand for portable and wearable healthcare devices, it is essential to create humid sensors that possess specific characteristics, such as being small, having a fast response time, being easily integrated, and having a low manufacturing cost. Additionally, incorporating machine learning (for example, deep learning methods like convolutional neural networks) can enhance signal processing and recognition accuracy while maintaining reliability.

Assistant Professors Kwai Hei Li’s and Yuanjing Lin’s research groups from the School of Microelectronics at the Southern University of Science and Technology (SUSTech) have recently proposed a fully integrated patch for wireless intelligent respiratory monitoring.

Their research work, entitled “Fully Integrated Patch Based on Lamellar Porous Film Assisted GaN Optopairs for Wireless Intelligent Respiratory Monitoring”, has been published in Nano Letters, a top journal covering all aspects of nanoscience and nanotechnology and their subdisciplines.

Figure 1. (a) Schematic diagram of the integrated patch. (b) Optical image of the sensing patch and a magnified view showing the humidity-sensing device. (c) Schematic illustration of the operation of the patch for wireless intelligent respiratory monitoring.

In this work, a flexible patch that is fully integrated for wireless smart monitoring of respiratory function was designed. This patch was based on a layered porous membrane and functionalized GaN optoelectronic devices. The GaN device at the sub-millimeter scale has a high sensitivity of up to 13.2 nA/%RH for relative humidity between 2-70% and 61.5 nA/%RH for relative humidity between 70-90%. An integrated wireless data transmission module and machine learning, utilizing a one-dimensional convolutional neural network, achieves the classification of seven respiratory modes with an overall accuracy greater than 96%. The fully integrated smart sensing patch can be used for non-invasive, real-time health monitoring applications.

Zecong Liu and Junjie Su, master’s students from the School of Microelectronics at SUSTech, are the co-first authors of this paper. Assistant Professors Kwai Hei Li and Yuanjing Lin are the co-corresponding authors.

This work was supported by the National Natural Science Foundation of China (NSFC), Shenzhen Natural Science Foundation Stability Support Program Project, Shenzhen Fundamental Research Program, and the SUSTech Grant.

Paper link: https://pubs.acs.org/doi/10.1021/acs.nanolett.3c02071

To read all stories about SUSTech science, subscribe to the monthly SUSTech Newsletter.