Scholarlink

<p>This study presents an innovative hybrid plasmonic biosensor design for non-invasive glucose detection in urine samples. Through optimization of geometric parameters and electrical properties, the sensor demonstrates exceptional performance with maximum absorption of 1.589 at 80° incident angle and graphene chemical potential of 0.9&#xa0;eV. The device exhibits a competitive sensitivity of 1000&#xa0;GHz/RIU, matching or exceeding existing designs, with optimal operation at 0.321 THz where maximum field confinement occurs. The sensor shows a frequency tuning range of 25&#xa0;GHz (0.33 THz to 0.305 THz) for glucose detection, with figure of merit values ranging from 58.82 to 9.80 RIU⁻<sup>1</sup>. On the other the integration of machine learning demonstrates the remarkable performance with the ability of cutting down simulation time and resources. This multi-material approach leverages the complementary advantages of each component while mitigating individual material limitations, offering a promising solution for point-of-care glucose monitoring applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Next-Generation Hybrid Multi-Material Surface Plasmon Resonance Biosensor for Non-Invasive Glucose Detection with Machine Learning Optimization

  • 期刊论文

  • 发表日期：

    2025年7月15日

查看全文
  • Ponlatha S,
  • Gomathy V,
  • Arun Kumar U,
  • Taha Sheheryar

摘要

This study presents an innovative hybrid plasmonic biosensor design for non-invasive glucose detection in urine samples. Through optimization of geometric parameters and electrical properties, the sensor demonstrates exceptional performance with maximum absorption of 1.589 at 80° incident angle and graphene chemical potential of 0.9 eV. The device exhibits a competitive sensitivity of 1000 GHz/RIU, matching or exceeding existing designs, with optimal operation at 0.321 THz where maximum field confinement occurs. The sensor shows a frequency tuning range of 25 GHz (0.33 THz to 0.305 THz) for glucose detection, with figure of merit values ranging from 58.82 to 9.80 RIU⁻1. On the other the integration of machine learning demonstrates the remarkable performance with the ability of cutting down simulation time and resources. This multi-material approach leverages the complementary advantages of each component while mitigating individual material limitations, offering a promising solution for point-of-care glucose monitoring applications.