Scholarlink

<p>The identification and diagnosis of brain tumors present considerable challenges in medical imaging, requiring innovative diagnostic techniques that enable precise and minimally invasive examinations. This study proposes a metasurfaces design for early brain tumor detection. The design incorporates graphene, gold, and methylammonium lead halide on SiO<sub>2</sub> substrate, with a T-shaped resonator structure that is optimized through detailed numerical simulations. Leveraging terahertz technology, the sensor depicts remarkable performance. Advanced simulations conducted using COMSOL Multiphysics exemplifies exceptional transmission performance in the THz regime. The sensor exhibits remarkable sensing properties, including 1538 GHzRIU⁻<sup>1</sup> and 29.586 RIU⁻<sup>1</sup> as the Optimal sensitivity and Figure of Merit respectively. Additionally, the sensor exemplifies 5.1 × 10⁻<sup>2</sup> as the minimum detection limit, offering exceptional resolution. To enhance the predictive performance, an XGBoost machine learning regression model is applied, demonstrating a remarkable performance with degree of correlation spanning from 98 to 100% across varying incident angles and refractive index changes. The design demonstrates significant advancement in non-invasive brain tumor detection, providing a promising avenue for early diagnosis and potential clinical 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

A Graphene-Based Surface Plasmon Resonance Metasurfaces Terahertz Sensor for Early Brain Tumor Detection with Machine Learning Optimization

  • 期刊论文

  • 发表日期：

    2025年3月29日

查看全文
  • Jacob Wekalao,
  • Yahya Ali Abdelrahman Ali,
  • Taoufik Saidani,
  • Shobhit K. Patel,
  • Abdulkarem H. M. Almawgani,
  • Basim Ahmad Alabsi

摘要

The identification and diagnosis of brain tumors present considerable challenges in medical imaging, requiring innovative diagnostic techniques that enable precise and minimally invasive examinations. This study proposes a metasurfaces design for early brain tumor detection. The design incorporates graphene, gold, and methylammonium lead halide on SiO2 substrate, with a T-shaped resonator structure that is optimized through detailed numerical simulations. Leveraging terahertz technology, the sensor depicts remarkable performance. Advanced simulations conducted using COMSOL Multiphysics exemplifies exceptional transmission performance in the THz regime. The sensor exhibits remarkable sensing properties, including 1538 GHzRIU⁻1 and 29.586 RIU⁻1 as the Optimal sensitivity and Figure of Merit respectively. Additionally, the sensor exemplifies 5.1 × 10⁻2 as the minimum detection limit, offering exceptional resolution. To enhance the predictive performance, an XGBoost machine learning regression model is applied, demonstrating a remarkable performance with degree of correlation spanning from 98 to 100% across varying incident angles and refractive index changes. The design demonstrates significant advancement in non-invasive brain tumor detection, providing a promising avenue for early diagnosis and potential clinical applications.