Towards Comprehensive Health Monitoring: Novel Fabrication and Applications of Intelligent Wearable Sweat Sensors

Abstract

In the pursuit of non-invasive and continuous health monitoring, flexible and wearable sensors have emerged as pivotal technologies, particularly for the detection of biomarkers in human sweat. Despite significant advancements, challenges persist in creating sensors that are both efficient in detection and comfortable for long-term wear. These challenges include complex and costly manufacturing processes, user discomfort due to poor breathability and flexibility, and difficulties in achieving real-time, accurate data transmission necessary for effective health management.

To address these issues, this research presents a series of innovative solutions in the development of flexible and wearable sensors. Firstly, a non-enzymatic flexible glucose sensor based on CuO/CaTiO₃ was developed using thermal transfer printing technology. This method offers low-cost, rapid, and small-batch fabrication. The sensor leverages the excellent electrocatalytic activity of CuO for glucose oxidation and the large specific surface area and good biocompatibility of CaTiO₃ to enhance electron transfer rates and broaden the linear detection range. A specially designed NaOH/Nafion/PEO blend film creates a consistent alkaline environment, eliminating the need for pretreatment. The sensor demonstrated high sensitivity (487.3 μA mM⁻¹ cm⁻²), a low detection limit (0.75 μM), a wide dynamic linear range (0.01 mM to 2 mM), and rapid response time (<0.1 s), along with excellent flexibility, stability, and biocompatibility in detecting glucose concentrations in human sweat.

Building upon this foundation, a novel bionic skin sensor was introduced, featuring a bilayer unidirectional moisture transport nanomembrane for enhanced comfort and real-time health monitoring. The bilayer design comprises an inner layer that facilitates sweat absorption and breathability and an outer layer that provides waterproofing and radiative cooling. This bionic skin was also manufactured using the thermal transfer process, suitable for small-scale mass production. It is capable of monitoring multiple biomarkers in sweat, including glucose, lactic acid, uric acid, pH, temperature, and skin impedance. Integration with the Continuous Analyte Monitoring with Real-time Engagement (CARE) system enables real-time data transmission and processing, enhancing the practical utility of the sensor in health monitoring applications.

Advancing the concept further, a sandwich-structured bionic skin with dual-layer nanofiber was developed, integrating multifunctional detection capabilities with deep learning wireless data transmission. The fabrication process was optimized using a zone-specific localized compression technique within the thermal transfer printing method, significantly reducing production time and cost. The dual-layer structure consists of an inner polyacrylonitrile (PAN/F127) layer for improved sweat absorption and an outer polydimethylsiloxane (PVB/PDMS) layer that offers waterproofing, radiative cooling, and antifouling properties. This design ensures exceptional comfort and mimics the mechanical properties of natural skin while maintaining enzymatic activity through temperature regulation. The sensor demonstrated high sensitivity and selectivity for each analyte, with long-term stability retaining over 84% of its initial performance after nine days. The integration of a redesigned Deep Learning Continuous Analyte Monitoring with Real-time Engagement (D-CARE) system, featuring a flexible circuit board and a deep learning attenuation correction algorithm, facilitates precise and real-time health data collection and wireless transmission. In situ testing during exercise confirmed the device's capability for continuous and accurate physiological monitoring.

Collectively, this thesis contributes significant advancements in the field of wearable health monitoring devices. By addressing key limitations related to manufacturing complexity, user comfort, and data transmission, the developed sensors offer reliable, real-time health monitoring in a comfortable and user-friendly format. These innovations hold promise for next-generation personalized healthcare, enhancing the accessibility and effectiveness of wearable technologies in continuous health management.

Date of Award	22 Dec 2025
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Chenjie XU (Supervisor), Jinlian HU (Supervisor) & Youhua Tan (External Co-Supervisor)