Fusing Digital-Breast-Tomosynthesis and Nearfield-Radar-Imaging Information for a Breast Cancer Detection Algorithm

Abstract

Breast cancer is the second leading cause of cancer death among women, with a rate of approximately 22.2 cases per 100,000. Conventional two-dimensional (2-D) mammography is currently the most effective screening tool for the early detection of breast cancer. Digital Breast Tomosynthesis (DBT) outperforms Mammography by allowing high in-plane resolution volumetric imaging, but it still suffers from the limited radiological contrast between cancerous and healthy tissues. We present a computational simulation of a hybrid Digital-Breast-Tomosynthesis (DBT) and microwave Nearfield Radar Imaging (NRI). The NRI measurements would be simultaneously collected using microwave antennas located on the surface of the breast to enable exact registration of the two imaging modalities. Since successful detection requires high sensitivity and low noise level, an accurate NRI simulation technique is needed. The DBT image is used to obtain a background distribution of the heterogeneous tissues, which become inputs to the NRI algorithm to successfully detect lesions in the breast a lesion and normal glandular/fibro-connective tissues. Using Finite Difference in the Frequency Domain simulations and real DBT data, we show that it should be possible to distinguish between cancerous and normal fibroglandular tissue, in spite of the low dielectric contrast between them.