<p>Cancer remains a leading cause of mortality worldwide, necessitating the development of rapid, cost-effective, and highly sensitive diagnostic technologies. In this study, a surface plasmon resonance (SPR) biosensor based on a TaSe<sub>2</sub>/Si/Ag multilayer structure in the Kretschmann configuration is proposed for label-free detection of colorectal cancer biomarkers. The incorporation of two-dimensional TaSe<sub>2</sub> enhances plasmonic coupling and near-field confinement at the metal–dielectric interface. Finite element method (FEM) simulations using COMSOL Multiphysics were employed to optimize the layer thicknesses, resulting in an optimal configuration comprising 41&#xa0;nm Ag, 3&#xa0;nm TaSe<sub>2</sub>, and 1.2&#xa0;nm Si. The optimized structure achieves a minimum reflectance of 0.022% with significant electric field enhancement. The sensor operates over a refractive index range of 1.36–1.401 RIU, enabling detection of biologically relevant variations. A maximum angular sensitivity of 500°/RIU, figure of merit of 103.37 RIU<sup>−1</sup>, quality factor of 17.78, detection accuracy of 0.207<sup>−1</sup>, and detection limit of 0.011 RIU are achieved. A strong linear relationship (R<sup>2</sup> = 0.99142) between resonance angle and refractive index confirms reliable sensing performance. Furthermore, machine learning models were employed to predict sensor responses, achieving prediction accuracy exceeding 99.6% (R<sup>2</sup> &gt; 0.996) while significantly reducing computational time. The proposed integration of advanced 2D materials and machine learning provides a promising platform for early-stage cancer screening and early&#xa0;detection and for point-of-care diagnostics.</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

Computational modeling and machine learning–assisted optimization of a TaSe2/Si/Ag multilayer SPR biosensor for high-sensitivity cancer biomarker screening and detection

  • M. S. Sivagama Sundari,
  • Azath Mubarakali,
  • U. Arun Kumar,
  • Shoeib Amin Banday

摘要

Cancer remains a leading cause of mortality worldwide, necessitating the development of rapid, cost-effective, and highly sensitive diagnostic technologies. In this study, a surface plasmon resonance (SPR) biosensor based on a TaSe2/Si/Ag multilayer structure in the Kretschmann configuration is proposed for label-free detection of colorectal cancer biomarkers. The incorporation of two-dimensional TaSe2 enhances plasmonic coupling and near-field confinement at the metal–dielectric interface. Finite element method (FEM) simulations using COMSOL Multiphysics were employed to optimize the layer thicknesses, resulting in an optimal configuration comprising 41 nm Ag, 3 nm TaSe2, and 1.2 nm Si. The optimized structure achieves a minimum reflectance of 0.022% with significant electric field enhancement. The sensor operates over a refractive index range of 1.36–1.401 RIU, enabling detection of biologically relevant variations. A maximum angular sensitivity of 500°/RIU, figure of merit of 103.37 RIU−1, quality factor of 17.78, detection accuracy of 0.207−1, and detection limit of 0.011 RIU are achieved. A strong linear relationship (R2 = 0.99142) between resonance angle and refractive index confirms reliable sensing performance. Furthermore, machine learning models were employed to predict sensor responses, achieving prediction accuracy exceeding 99.6% (R2 > 0.996) while significantly reducing computational time. The proposed integration of advanced 2D materials and machine learning provides a promising platform for early-stage cancer screening and early detection and for point-of-care diagnostics.