Abstract Integration of an adaptive finite element method (AFEM) with a conventional least squares method has been presented. As a 3D full-wave forward solver, CST Microwave Studio has been used to model and extract both electric field distribution in the region of interest (ROI) and S-parameters of a circular array consisting of 16 monopole antennas. The data has then been fed into a differential inversion scheme to get a qualitative indicator of how the temperature distribution evolves over a course of the cooling process of a heated object. Different regularization techniques within the Tikhonov framework are also discussed, and a balancing principle for optimal choice of the regularization parameter was used to improve the image reconstruction quality of every 2D slice of the final image. Targets are successfully imaged via proposed numerical methods.

中文翻译：

---

微波测温在无创体温监测中的潜在应用

摘要 提出了自适应有限元方法 (AFEM) 与传统最小二乘法的集成。作为 3D 全波正向求解器，CST Microwave Studio 已被用于建模和提取感兴趣区域 (ROI) 中的电场分布和由 16 个单极天线组成的圆形阵列的 S 参数。然后将数据输入差分反演方案，以获得温度分布如何在加热物体的冷却过程中演变的定性指标。还讨论了 Tikhonov 框架内的不同正则化技术，并使用了最佳选择正则化参数的平衡原则来提高最终图像的每个 2D 切片的图像重建质量。