The robustness of HGA/FDTD in the presence of noise for microwave breast cancer detection

In this paper, the robustness of the microwave tomography method for breast cancer imaging based on finite difference time domain (FDTD) and hybrid genetic algorithm (HGA) for different levels of noise is investigated. Results based on the proposed method, using MRI data from an actual patient, are presented. These results indicate the algorithm performs well in the case of simulation measurements contaminated by various levels of additive white Gaussian noise (up to 15% of signal strength). Noise levels above this value inhibit the efficacy of the method.