Scholarlink

<p>This study proposes an image acquisition method using scattering parameters based on the evaluation of tissue information in the reflection coefficient curves. An antenna with breast tissue in its mid-field was simulated at millimeter wave frequencies, consecutively for varying tissue types, sizes, and locations. The resulting 3468 reflection coefficient curves |<i>S</i>11| were assessed by Mutual Information and Machine Learning (ML) algorithms. The accuracy of prediction models achieved %96.4 in tissue size, and regression models localized tissues with errors less than their sizes. ML models extracted the tissue information from an |<i>S</i>11| curve, however, the extraction process varied from model to model and was not defined. Considering that the interference phenomenon identifies the relation between tissue and |<i>S</i>11| curve, an interference pattern was created using the |<i>S</i>11| curve. Then, the pattern was assumed as a coefficient matrix of a system of linear equations, where the unknowns were tissue vectors in the radiation field. The solution of this system provided the tissues’ contributions to |<i>S</i>11|. Therefore, a reflection pattern and |<i>S</i>11| curves made image acquisition attainable, and a distinctive method, Reflection Coefficient-based Linear Tomographic Imaging (RC-LTI), was presented for breast cancer detection.</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 Linear system of reflection coefficients for tomographic imaging of breast cancer

  • 期刊论文

  • 发表日期：

    2024年12月31日

查看全文
  • Nurhan Güneş,
  • İremnur Duru,
  • Timuçin Emre Tabaru

摘要

This study proposes an image acquisition method using scattering parameters based on the evaluation of tissue information in the reflection coefficient curves. An antenna with breast tissue in its mid-field was simulated at millimeter wave frequencies, consecutively for varying tissue types, sizes, and locations. The resulting 3468 reflection coefficient curves |S11| were assessed by Mutual Information and Machine Learning (ML) algorithms. The accuracy of prediction models achieved %96.4 in tissue size, and regression models localized tissues with errors less than their sizes. ML models extracted the tissue information from an |S11| curve, however, the extraction process varied from model to model and was not defined. Considering that the interference phenomenon identifies the relation between tissue and |S11| curve, an interference pattern was created using the |S11| curve. Then, the pattern was assumed as a coefficient matrix of a system of linear equations, where the unknowns were tissue vectors in the radiation field. The solution of this system provided the tissues’ contributions to |S11|. Therefore, a reflection pattern and |S11| curves made image acquisition attainable, and a distinctive method, Reflection Coefficient-based Linear Tomographic Imaging (RC-LTI), was presented for breast cancer detection.