Testing for Convergence in Microwave Breast Tumour Detection

Finding an exact solution for some critical parameters of an internal object using the measured data is a challenging task in practical applications. The inverse method has to deal with two major situations. One is the noise which can be included in the measurement data and the other is the lack of information about the unknowns. When finding solutions to the inverse problem analytically, we have information about the exact values of the unknowns, but in practical applications this information is available only with certain limitations. This paper presents a scanning method of testing for convergence to detect an internal object using microwave measurements. Data acquisition is based on the system of a microwave scattering technique considering both forward and backscattered wave-fronts. A computational scanner is used to find solutions to both forward and inverse problems. Modelling of this application system has been aimed towards the in-vivo detection of a breast tumour, in particular, but this can be used for detecting an unknown object inside a host material in general. The internal object inside the model has been considered to be a sphere and the subsequent scattering problem has been solved using a three-dimensional coordinate system. The nonlinear inverse scattering problem is solved to find few unknowns so that the location and the size of the internal object like a breast tumour can be found using few microwave measurements.