Microwave Tomography. Microwave tomography techniques have been studied by several research groups [5, 22– 47]. Xxxx. Xxxxxxx and Xxxx. Xxxxxx from Dartmouth College in the United States are one of the representative research groups. They have studied microwave tomography breast imaging since 1990s [48–51]. And they proposed an iterative reconstruction algorithm by obtaining dielectric properties in a 2D lossy medium in 1995. The goal of microwave tomography is to recover the profile using the inverse problem of the dielectric properties of the breast. Microwave tomography uses an inverse scatter- ing method to develop a diagnostic image of the breast. Inverse scattering uses scattering signals, including diffraction from objects. It creates a map of permittivity and conduc- tivity through inversion of those signals [52]. However, the inverse problem takes an increased amount of time, because the calculation process is complicated. Furthermore, a non-linear inverse scattering problem must be solved, and iterative image reconstruction algorithms are usually required to obtain a solution. In general, these ill-posed inverse scattering approaches suffer from non-uniqueness and require regularisation in order to achieve convergence to a meaningful solution [29, 53]. The Dartmouth group, for the first time, have developed a clinical prototype for active MWI of the breast [5]. They developed a 32-channel data acquisition system operating at a frequency range of 500 MHz to 3 GHz [49] to obtain data from a clinical prototype exam. The clinical prototype illuminates the breast with 16 monopole antennas that operate in the 300 MHz to 1 GHz frequency range. The clinical exam was conducted with five women, and the total acquisition time was 10–15 minutes per breast. The system measured and obtained data at seven different array heights for seven different frequencies at each array position [52]. King’s College London is one institute that has been experimenting on different antenna designs for microwave tomography [44, 46]. They have also proposed new imaging algorithms, such as an adaptive thresholding iterative shrinkage algorithm to the linear inversion at each iteration of the distorted Born iterative method (DBIM) [54]. Other groups are researching on reducing high computational demands due to a number of iterations and discretisation of the object image for accuracy. Xx et al. at the University of Manitoba proposed an iterative process that involves two algorithms: finite-differenc...
