On the Performance Revision of a Wearable Antenna Sensor for Glucose Detection Utilizing Artificial Neural Networks
Main Article Content
Abstract
The study "Performance Revision of a Wearable Antenna Sensor for Glucose Detection Utilizing Artificial Neural Networks" explores machine learning techniques to improve the accuracy and responsiveness of a wearable antenna sensor for glucose detection. The results demonstrate significant improvements in the sensor's performance, particularly in real-time glucose monitoring. By fine-tuning the neural network architecture, the researchers achieved a higher degree of precision while minimizing false readings, enhancing the accuracy of glucose measurements and paving the way for more customized diabetes management strategies. The findings suggest that integrating additional data sources, such as patient activity levels and dietary habits, could further refine the predictive capabilities of the system. Future work will focus on integrating these advancements into a compact, wearable format, ensuring user comfort and accessibility. Additionally, exploring the potential for remote monitoring features could further empower individuals in managing their health proactively. The study also addresses the issue of non-linearity due to diffraction effects from different layers in microwave resonators. The proposed antenna designs are based on a low-cost, highly-sensitive microwave antenna for identifying different liquid samples through monitoring the variation in S21 magnitude. The antenna's accuracy is improved through a back loop trace technique, eliminating nonlinear effects due to multi-layer diffractions. An analytical model based on circuit theory is suggested for the proposed antenna operation, and the results are validated numerically using the Computer Simulation Technology Microwave studio package.