Measurement uncertainties in differential radar applied to breast imaging

Abstract

In this work, we evaluate the measurement uncertainty and its consequences on a time-domain microwave breast imaging system. Our radar system contains a 16-element multistatic sensor array, and is used to generate microwave images of the breast using a differential method. We examine, for the first time with such a system, uncertainty due to sources of horizontal and vertical noise individually and together to determine their impact on image quality and tumor detection. It is found that for a time-domain radar system using differential imaging, horizontal noise is significantly more detrimental than vertical noise, and must be compensated for in order to have successful imaging of breast tumors. We show an example of a reconstructed breast image before and after compensation for both types of noise and highlight the necessity of such compensation.