Related Experiment Video
Updated: Aug 9, 2026
11:04
NiO Nanoflowers for Non-Enzymatic Amperometric Detection of Glucose
Published on: December 30, 2025
High-Sensitivity Non-Invasive Microwave Glucose Sensor with ZnO/CNT Composite Optimized by Deep Learning for Wearable
Jiaxu Liu1, Zhao Yao1, Qingzhou Wang1,2
1Shandong Key Laboratory of Micro-nano Packaging and System Integration, College of Electronic and Information, Qingdao University, Qingdao 266071, China.
ACS Omega
|August 8, 2026
Summary
A new microwave sensor uses zinc oxide/carbon nanotubes for noninvasive diabetes monitoring. A convolutional neural network (CNN) achieves high accuracy (R² = 0.98) in predicting blood glucose levels.
Area of Science:
- Biomedical Engineering
- Materials Science
- Electromagnetics
Background:
- Diabetes prevalence is rising globally, necessitating improved detection and monitoring.
- Conventional glucose monitoring methods are often invasive and complex.
- Noninvasive monitoring offers a promising alternative for diabetes management.
Purpose of the Study:
- To develop a novel microwave sensor for noninvasive blood glucose detection.
- To enhance sensor sensitivity and response using a zinc oxide/carbon nanotube composite.
- To validate the sensor's performance using simulations and human volunteer testing.
Main Methods:
- Fabrication of a microwave sensor integrating electromagnetic coupling with a ZnO/CNT composite.
- Utilized Sim4Life electromagnetic simulator for human tissue modeling and validation.
- Applied a convolutional neural network (CNN) for data analysis and glucose level prediction.
Main Results:
- The ZnO/CNT composite significantly improved sensor sensitivity and response.
- Simulations validated the sensor's detection capability at 3 GHz.
- The CNN model achieved high-precision noninvasive glucose prediction (R² = 0.98).
Conclusions:
- The developed microwave sensor offers an efficient solution for noninvasive glucose monitoring.
- This technology shows significant potential for early diabetes screening and continuous monitoring.
- The study highlights the feasibility of advanced sensing and AI for wearable diabetes devices.
