A deep learning method is applied to modelling electromagnetic (EM) scattering for microwave breast imaging (MBI). The neural network (NN) accepts 2D dielectric breast maps at 3 GHz and produces scattered-field data on an antenna array composed of 24 transmitters and 24 receivers. The NN was trained by 18,000 synthetic digital breast phantoms generated by generative adversarial network (GAN), and the scattered-field data pre-calculated by method of moments (MOM). Validation was performed by comparing the 2,000 NN-produced datasets isolated from the training data with the data computed by MOM. Finally, data generated by NN and MOM were used for image reconstruction. The reconstruction demonstrated that errors caused by NN would not significantly affect the image result. But the computational speed of NN was nearly 104 times faster than the MOM, indicating that deep learning has the potential to be considered as a fast tool for EM scattering computation.
Keywords: Computational electromagnetics; convolutional neural network; deep learning; microwave imaging.