Interpreting Artificial Neural Networks for Microwave Detection of Breast Cancer

摘要

This letter demonstrates the applicability of artificial neural networks (ANNs) to breast cancer detection. A simplified model of the breast containing a tumor was used to determine the scattering of electromagnetic waves in the microwave band. This data was used to train an ANN which was then tested on new data to predict the presence or absence of a tumor. Synthetic variation (noise) is added to the data to realistically model the breast.