Temperature-Induced Contrast Enhancement for Radar-Based Breast Tumor Detection at K-Band Using Tissue Mimicking Phantoms

Conventional approaches for microwave breast tumor detection are limited by the imaging resolution due to the low operating frequency. The objective of this work is to provide a proof of concept for radar-based detection of breast tumors in K-band using the temperature-dependent permittivity of the tissue for contrast enhancement. The innovation of this work is given by i) investigating higher microwave frequencies for breast cancer diagnostics and improved resolution; ii) exploiting variations in tissue temperature as a non-invasive approach for contrast-induced radar imaging eliminating the need for contrast agents such as nanoparticles; iii) using a well-defined setup with the breast compressed similar to mammography; iv) eliminating the need for coupling liquid through the usage of ultra-wideband bow-tie antennas operating from 16.55 to 40 GHz for a reflection coefficient lower than −10 dB; v) validating the experimental findings through numerical modelling. The experimental setup in this work consists of a single-pixel transmission setup with the antennas placed in a 3D printed container. Two different tissue mimicking phantoms have been studied that both model the temperature-dependent permittivity of biological tissue. The first phantom represents homogeneous fatty tissue properties and the second phantom simulates fatty tissue with a tumor inclusion. A uniform phantom warming is realized through a water bath combined with a continuous monitoring of the phantoms temperature. We show that a homogeneous phantom without tumor can be distinguished from a heterogeneous phantom with tumor in the temperature range of 28
$^\circ$
C to 38
$^\circ$
C.