This paper aims a new implementation of the popular distorted Born iterative method (DBIM) that could effectively solve the inverse scattering problem in microwave breast cancer imaging systems.
Instead of a tomographic reconstruction of the unknown permittivity profile of the breast, the proposed algorithm adopts a 3D approach. The algorithm iteratively solves the complex permittivity profile of 3D scatterers with method of moment-based volume integral equation solver. Instead of the surface meshing approach based on Rao−Wilton−Glisson basis function, a volume meshing is adopted here, and implemented on Schaubert−Wilton−Glisson (SWG) basis function. The imaging configuration consists of a group of Hertzian dipole antennas arranged on a hemispherical surface covering the breast. By applying the inverse solver algorithm on these scattered data sets, the complex permittivity profile of the 3D object under test is faithfully reconstructed.
The proposed algorithm is tested on a small cube scatterer, and a realistic human breast model. A noise analysis is also performed to assess the robustness of the algorithm under measurement noise. The electromagnetic radiation used in this method has a frequency of 2 GHz. The soft tissues in human breast guarantee the sufficient penetration depth required by the electromagnetic radiation for the complex permittivity reconstruction.
The novelty of this research is in tailoring DBIM onto a 3D full wave solver with SWG basis function to implement an accurate inverse solver.
