Dispersive Dielectric Properties of Breast Tumor Composition Detection Utilizing a Frequency Dependent Inverse Scattering Technique
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
UNIVERSITI MALAYSIA SARAWAK
Abstract
Description
Cancer is the major cause of morbidity and mortality worldwide and breast cancer remains the most common cancer and major cause of death among women worldwide since 1940’s. Frequency-Dependent Forward-Backward Time-Stepping (FD-FBTS) method is one of the inverse scattering techniques in time domain in which is applied to microwave imaging breast cancer detection by taking into account the dispersion effects of the human tissues. The FD-FBTS algorithm is implemented in C++ language executed in parallel computing by utilizing the Finite-Difference Time-Domain (FDTD) in time-domain form to reconstruct the microwave properties of the breast. Conjugate gradient (CG) method is integrated into the FD-FBTS algorithm to optimize the computational time and reconstructions. Dai Yuan CG method demonstrated simultaneous reconstructions for dispersive dielectric properties of the object or breast composition compared to others CG methods. To demonstrate the validity of the FD-FBTS algorithm, two-dimensional (2D) cases were assessed via numerical simulations. The FD-FBTS algorithm was first assessed with simple cases which is reconstructing simple circular object embedded within the region of interest (ROI) and extended to investigate on realistic numerical breast model which are heterogeneous and extremely dense realistic numerical breast model. The reconstructed images demonstrated the ability of FD-FBTS algorithm in estimating the location and dispersive dielectric properties of the embedded circular object as small as 4 mm radius of the circular object and embedded tumor as small as 8 mm diameter for both type realistic numerical breast models. The reconstruction results demonstrated that the FD-FBTS inverse scattering algorithm has the ability to detect and localize the presence of tumor accurately and provide evidence of FD-FBTS algorithm is applicable for object detection and tumor detection in various types of breast.