A multifaceted approach to enhancement of microwave breast images: Exploitation of magnetic contrast, use of a tissue-dependent mapping technique, and implementation of adaptive algorithmic stopping criteria

Loading...
Thumbnail Image
Date
2020-08
Authors
Kaye, Cameron
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Although microwave imaging (MWI) remains a promising future diagnostic tool in breast cancer detection and monitoring, its progress towards clinical application would benefit from improvement to its relatively low spatial and contrast resolution in delineating healthy and abnormal tissues. Research into multiple quality improvement methods for MWI reconstructions has thus been undertaken, through alterations of an existing finite-element contrast source inversion (FEM-CSI) algorithm that uses a time-harmonic, discontinuous Galerkin formulation of Maxwell's equations (DGM-CSI). Firstly, DGM-CSI has been employed to produce 2D images of the dielectric properties of synthetic breast models at multiple frequencies using a modified frequency-hopping technique that successively introduces alterations to the intermediate reconstructions of the complex-valued permittivity obtained at each of the individual frequencies. Using a tissue-dependent mapping technique, improvement in overall image quality is observed when the imaginary part of a reconstruction is modified to reflect the identical tissue geometry and probable tissue types obtained in its real part, and passed back to the algorithm as the initial guess between successive frequencies. Secondly, related increases in efficiency are noted when this same DGM-CSI algorithm is altered to employ a “frequency cycling” reconstruction technique combined with novel automated stopping criteria. Images obtained from the highest frequency data in a given frequency-hopping sequence are cycled back to the lowest frequency data as new initial guesses to restart multi-frequency inversions. The stopping criteria reduce the total number of algorithmic iterations by determining a suitable time to shift imaging frequencies and globally terminate reconstructions, based on a statistical analysis of a window of past iterations of data error using the two-sample Kolmogorov-Smirnov (K-S) test. Finally, contrast enhancement using “ferrofluid” colloidal suspensions of magnetic nanoparticles (MNPs) is introduced to MWI data collection and validated through a preliminary signal acquisition test using an external polarizing magnetic field (PMF) across a faceted metallic resonant enclosure. The technique of PMF-mediated modulation of the target MNPs’ ferromagnetic resonance properties is denoted herein as “ferromagnetic resonance imaging” (FRI). A two-stage imaging DGM-CSI methodology has been developed to recover both the relative permittivity and permeability of dielectric and magnetic targets for future imaging studies using FRI.
Description
Keywords
Microwave imaging, Ferromagnetic resonance, Magnetic nanoparticles, Inverse scattering, Applied electromagnetics, Resonant cavities, Breast imaging, Discontinuous Galerkin method, Contrast source inversion, Stopping criteria, Kolmogorov-Smirnov test
Citation
C. Kaye, I. Jeffrey, and J. LoVetri, "Improvement of Multi-Frequency Microwave Breast Imaging through Frequency Cycling and Tissue-Dependent Mapping," IEEE Transactions on Antennas and Propagation, vol. 67, no. 11, pp. 7087-7096, Nov. 2019.
C. Kaye, I. Jeffrey, and J. LoVetri, "Novel Stopping Criteria for Optimization-Based Microwave Breast Imaging Algorithms," Journal of Imaging, vol. 5, no. 5, p. 55, May 2019.
C. Kaye, C. Gilmore and J. LoVetri, "Enhanced Detection of Magnetic Nanoparticles using a Novel Microwave Ferromagnetic Resonance Imaging System," in IEEE Transactions on Biomedical Engineering (Early Access), doi: 10.1109/TBME.2020.3019716.
C. Kaye, I. Jeffrey, and J. LoVetri, "Two-stage reconstruction of complex dielectric permittivity and magnetic permeability for biomedical microwave imaging employing magnetic contrast agents," in Proceedings of the 4th Advanced Electromagnetics Symposium (AES), Malaga, Spain, Jul. 2016, pp. 40-42.