Scholarlink

<p>Traditional amino acid detection methods exhibit significant limitations including high costs, lengthy analysis times, and lack of portability. This study presents a metamaterial-enhanced graphene biosensor integrated with artificial intelligence for highly sensitive amino acid detection in the terahertz frequency range. Performance evaluation across 0.1–1 THz demonstrated exceptional sensitivity of 667GHzRIU<sup>−1</sup>, with transmittance values of 48.610–50.176% at resonant frequencies of 0.546–0.582 THz. Optimal performance occurred at 0.311 THz and 0.928 THz with minimal absorption losses. Random Forest regression algorithms achieved perfect R<sup>2</sup> values (100%) for test cases within the 0.1–0.4 range, with R<sup>2</sup> values improving from 91 to 100% as polynomial degrees increased from 2 to 10. The sensor demonstrates superior performance compared to existing solutions, offering a cost-effective, rapid, and highly accurate platform for amino acid detection with applications in clinical diagnostics, pharmaceutical development, and biotechnology.</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-Sensitivity Amino Acid Sensing Using Machine Learning-Optimized Graphene-Gold-Silver Metasurface-Based Surface Plasmon Resonance Biosensor in the Terahertz Regime

  • 期刊论文

  • 发表日期：

    2025年6月12日

查看全文
  • P. Umaeswari,
  • Jacob Wekalao,
  • Kumaravel Kaliaperumal

摘要

Traditional amino acid detection methods exhibit significant limitations including high costs, lengthy analysis times, and lack of portability. This study presents a metamaterial-enhanced graphene biosensor integrated with artificial intelligence for highly sensitive amino acid detection in the terahertz frequency range. Performance evaluation across 0.1–1 THz demonstrated exceptional sensitivity of 667GHzRIU−1, with transmittance values of 48.610–50.176% at resonant frequencies of 0.546–0.582 THz. Optimal performance occurred at 0.311 THz and 0.928 THz with minimal absorption losses. Random Forest regression algorithms achieved perfect R2 values (100%) for test cases within the 0.1–0.4 range, with R2 values improving from 91 to 100% as polynomial degrees increased from 2 to 10. The sensor demonstrates superior performance compared to existing solutions, offering a cost-effective, rapid, and highly accurate platform for amino acid detection with applications in clinical diagnostics, pharmaceutical development, and biotechnology.