Noninvasive Microwave-Multielement Sensor for Breast Phantoms Analysis and Tumor Detection
Madan Kumar Sharma, Satya P. Singh, Paarth Badola, Mithilesh Kumar, J. P. Saini, A. Lay-Ekuakille
Abstract
This article proposes a microwave sensor for breast phantom analysis and tumor detection. The sensor is developed on the small-size ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$25\times25$ </tex-math></inline-formula> mm2) low-cost RT/duroid 5880 substrate. The structure of the proposed sensor is made up of four sensory elements. The sensor performance is experimentally validated by the evaluation of the S-parameters and radiation patterns. The proposed sensor has a 2–30-GHz-wide operating bandwidth, lower mutual coupling ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{21}} \le -30$ </tex-math></inline-formula> dB) between the elements, and offered a desirable directional radiation pattern. Four types of breast phantoms are mimicked: normal phantom, phantom with 3-, 5-, and 10-mm tumor. Developed phantoms have experimented with the proposed sensor and tumor identification is carried out. Measurement of the S-parameters, i.e., reflection coefficients ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{11}}$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{22}}$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{33}}$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{44}}$ </tex-math></inline-formula> ) and mutual coupling coefficient ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{21}}$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{31}}$ </tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${S} _{{41}}$ </tex-math></inline-formula> ) with and without tumor phantom is executed. A large difference in the peak value of the S-parameters results for normal and malignant phantoms is observed over the 2–20-GHz frequency range. Reflection results from data are further processed through principal component analysis (PCA) to better visualize the difference between normal and malignant phantoms. Obtained PCA results exhibited a significant difference between normal and malignant phantoms. Specific absorption rate (SAR) analysis is also performed to separate normal and malignant phantoms. The proposed approach is a noninvasive, low-cost modality and can analyze the phantoms for tumor detection.