Scholarlink

<p>This work reports a data-driven InSb/Si Tunnel FET based biosensor that has capabilities to detect breast cancer cell lines (MDA-MB-231, T47D, HS578T, and MCF-7) and non-tumorigenic (MCF-10&#xa0;A) simultaneously. The proposed work focuses upon sequential-bias-reconfiguration of the Tunnel FET. The device demonstrated a high I<sub>ON</sub> of 10<sup>−4</sup> A/µm and threshold voltage of 0.026 to 0.029&#xa0;V for the breast cancer cell lines. To address computational efficiency for practical applications, the algorithmic framework is further refined through feature reduction technique to balance the performances between complexity of data acquisition, accuracy and data processing. By harnessing this machine-learning aided sensing device, we successfully demonstrated the potential of data-driven sensing devices in forthcoming health care applications. In this work, an artificial neural network model is used for predicting different parameters with an R<sup>2</sup> score of 0.99.</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

Early Detection of Breast Cancer Using Machine Learning Driven InSb/Si TFET Biosensor

  • 期刊论文

  • 发表日期：

    2025年10月17日

查看全文
  • Sukanta Kumar Swain,
  • Dillip Kumar Sahoo,
  • Kanhu Charan Bhuyan,
  • Shashi Kant Sharma

摘要

This work reports a data-driven InSb/Si Tunnel FET based biosensor that has capabilities to detect breast cancer cell lines (MDA-MB-231, T47D, HS578T, and MCF-7) and non-tumorigenic (MCF-10 A) simultaneously. The proposed work focuses upon sequential-bias-reconfiguration of the Tunnel FET. The device demonstrated a high ION of 10−4 A/µm and threshold voltage of 0.026 to 0.029 V for the breast cancer cell lines. To address computational efficiency for practical applications, the algorithmic framework is further refined through feature reduction technique to balance the performances between complexity of data acquisition, accuracy and data processing. By harnessing this machine-learning aided sensing device, we successfully demonstrated the potential of data-driven sensing devices in forthcoming health care applications. In this work, an artificial neural network model is used for predicting different parameters with an R2 score of 0.99.