权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Theoretical Study on High-Quality Visualization of Breast Cancer Using Microwaves

利用微波对乳腺癌进行高质量可视化的理论研究

基本信息

批准号：
13680943
负责人：
TANAKA Mitsuru
金额：
$ 1.47万
依托单位：
Oita University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2001
资助国家：
日本
起止时间：
2001 至 2003
项目状态：
已结题

项目摘要

It is the purpose of this research to provide high-quality visualization algorithms of imaging strongly scattering objects such as a breast cancer. The research results are summarized as follows.1.In the analysis of frequency-domain inverse scattering problems, we employ the genetic algorithm (GA), the conjugate gradient method, and the Levenberg-Marquardt method for the reconstruction of relative permittivities of objects. On the other hand, the GA and the conjugate gradient method are used in time-domain inverse scattering problems. Both the frequency-and time-domain inverse scattering problems are reduced to nonlinear optimization problems. Then we can derive efficient visualization algorithms effective in strongly scattering objects. Note that the GA is very useful for a one-dimensional inverse scattering problem.2.In frequency-domain inverse scattering problems, the convergence speed for the multifrequency case is much faster than that for the single-frequency case. The optimal re … More gularization parameter is determined by minimizing the absolute value of the radius of curvature of the generalized cross-validation (GCV) function. This method provides successful reconstructions with the property of much faster convergence than the conventional GCV method. To accelerate the visualization algorithm, we propose a new regularization method where the regularization term is dynamically changed with the residual error in the scattered electric field.3.The effect of an incident pulse on the accuracy of image reconstruction in time-domain inverse scattering is also studied. It is confirmed that the reconstruction accuracy can be considerably improved by selecting the wave form of an incident pulse. Especially, we can obtain good reconstructions by using the monocycle pulse.It is concluded from the research results that the visualization algorithms proposed have the properties of fast-convergence and high-resolution. Then these algorithms can be applied to the imaging of a breast cancer. Less

本研究的目的是提供高质量的可视化算法成像强散射的对象，如乳腺癌。主要研究成果如下：1.在频域逆散射问题的分析中，分别采用了遗传算法、共轭梯度法和Levenberg-Marquardt法来重建目标的相对散射系数。另一方面，将遗传算法和共轭梯度法应用于时域逆散射问题。频域和时域逆散射问题都归结为非线性优化问题。然后，我们可以推导出有效的可视化算法，有效地在强散射的对象。2.在频域逆散射问题中，多频情形的收敛速度比单频情形快得多。最佳反应 ...更多信息通过最小化广义交叉验证（GCV）函数的曲率半径的绝对值来确定Gularization参数。这种方法提供了成功的重建与性能的更快的收敛速度比传统的GCV方法。为了加快可视化算法的速度，我们提出了一种新的正则化方法，该方法中正则化项随散射电场中的残差动态变化。3.研究了入射脉冲对时域逆散射图像重建精度的影响。它被证实，重建精度可以大大提高通过选择入射脉冲的波形。研究结果表明，本文提出的可视化算法具有收敛速度快、分辨率高的特点。然后，这些算法可以应用于乳腺癌的成像。少