Scholarlink

<p>This research introduces a biosensor engineered for high-sensitivity cancer detection with dual-mode sensing functionalities. The suggested sensor exhibits outstanding performance characteristics, with a maximum sensitivity of 1 THz/RIU, a figure of merit of 13.158 RIU<sup>−1</sup>, and a low detection limit of 0.150 RIU. The sensor demonstrates remarkable tunability via modification of graphene’s chemical potential (0.1–0.9&#xa0;eV) and sustains consistent performance over a range of incidence angles (0°-80°). Machine learning regression models, particularly random forest algorithms, were successfully integrated to enhance real-time data interpretation and diagnostic accuracy, achieving an R<sup>2</sup> score of 0.85 across multiple operational conditions. The linear relationship between resonance frequency and refractive index (R<sup>2</sup> = 0.95276) ensures reliable calibration and quantification capabilities. Fabrication employs standard micro/nano-fabrication techniques including electron beam lithography, chemical vapor deposition, and sequential material deposition. Compared to existing biosensors, the proposed design offers superior sensitivity over a broader refractive index range (1.360–1.401), making it suitable for multi-purpose cancer detection applications in resource-constrained environments.</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

High-Performance Dual-Mode Terahertz Biosensor for Cancer Detection via Graphene Modulation and Machine Learning Integration

  • 期刊论文

  • 发表日期：

    2025年11月8日

查看全文
  • N. Vithyalakshmi,
  • N. K. Anushkannan,
  • U. Arun Kumar,
  • Taha Sheheryar

摘要

This research introduces a biosensor engineered for high-sensitivity cancer detection with dual-mode sensing functionalities. The suggested sensor exhibits outstanding performance characteristics, with a maximum sensitivity of 1 THz/RIU, a figure of merit of 13.158 RIU−1, and a low detection limit of 0.150 RIU. The sensor demonstrates remarkable tunability via modification of graphene’s chemical potential (0.1–0.9 eV) and sustains consistent performance over a range of incidence angles (0°-80°). Machine learning regression models, particularly random forest algorithms, were successfully integrated to enhance real-time data interpretation and diagnostic accuracy, achieving an R2 score of 0.85 across multiple operational conditions. The linear relationship between resonance frequency and refractive index (R2 = 0.95276) ensures reliable calibration and quantification capabilities. Fabrication employs standard micro/nano-fabrication techniques including electron beam lithography, chemical vapor deposition, and sequential material deposition. Compared to existing biosensors, the proposed design offers superior sensitivity over a broader refractive index range (1.360–1.401), making it suitable for multi-purpose cancer detection applications in resource-constrained environments.