Scholarlink

<p>This study presents the intelligent optimization and performance prediction of a D-shaped photonic crystal fiber (PCF) biosensor enhanced with localized surface plasmon resonance (LSPR) for ultra-sensitive analyte detection. The sensor incorporates a plasmonic gold nanoparticle (AuNP) coating to improve sensitivity by leveraging the interaction between the core-guided mode and the surface plasmon mode (SPM) at the metal–dielectric interface. Using the finite element method (FEM) for modal analysis, critical parameters, including the D-shaped channel radius (<i>R</i><sub>d</sub>), gold nanoparticle radius (<i>R</i><sub>g</sub>), and the angle between nanoparticles (<i>δ</i>), were optimized through response surface methodology (RSM) with a Box–Behnken design (BBD). To further refine the sensor’s predictive capabilities, a multilayer perceptron (MLP) artificial neural network was trained to model its performance, allowing precise estimation of sensitivity metrics within the specific refractive index (RI) range of 1.31–1.37. The optimized biosensor achieved a remarkable spectral sensitivity of 30,000&#xa0;nm/RIU, an amplitude sensitivity of 200.292 RIU<sup>−1</sup>, and a resolution of 3.33 × 10<sup>–6</sup> RIU. The integration of artificial intelligence in the design and optimization process highlights a paradigm shift in biosensor engineering, offering a powerful approach for real-time performance prediction and enhancing detection precision in biomedical diagnostics and environmental 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

Artificial Intelligence Optimization of Au Nanoparticles-Based Photonic Crystal Fiber Biosensor

  • 期刊论文

  • 发表日期：

    2026年6月1日

查看全文
  • Sameh Kaziz,
  • Lamia Guedri-Knani,
  • Fraj Echouchene,
  • Cherif Dridi

摘要

This study presents the intelligent optimization and performance prediction of a D-shaped photonic crystal fiber (PCF) biosensor enhanced with localized surface plasmon resonance (LSPR) for ultra-sensitive analyte detection. The sensor incorporates a plasmonic gold nanoparticle (AuNP) coating to improve sensitivity by leveraging the interaction between the core-guided mode and the surface plasmon mode (SPM) at the metal–dielectric interface. Using the finite element method (FEM) for modal analysis, critical parameters, including the D-shaped channel radius (Rd), gold nanoparticle radius (Rg), and the angle between nanoparticles (δ), were optimized through response surface methodology (RSM) with a Box–Behnken design (BBD). To further refine the sensor’s predictive capabilities, a multilayer perceptron (MLP) artificial neural network was trained to model its performance, allowing precise estimation of sensitivity metrics within the specific refractive index (RI) range of 1.31–1.37. The optimized biosensor achieved a remarkable spectral sensitivity of 30,000 nm/RIU, an amplitude sensitivity of 200.292 RIU−1, and a resolution of 3.33 × 10–6 RIU. The integration of artificial intelligence in the design and optimization process highlights a paradigm shift in biosensor engineering, offering a powerful approach for real-time performance prediction and enhancing detection precision in biomedical diagnostics and environmental monitoring applications.