Microwave imaging (MWI) is a non-ionizing, non-invasive and an upcoming affordable medical imaging modality. Over the last few decades, MWI has invited active research towards biomedical imaging, with special focus on breast tumor detection. After long years of intense research and clinical trials, a breast tumour monitoring unit based on MWI is finally entering clinical imaging scenarios. In this manuscript, the vast literature in MWI to date has been consolidated, and an indetail study of the state-of-the-art for breast tumor detection has been presented. The hurdles faced during clinical trials are discussed, and their possible solutions and future directions for a fast transition into clinical imaging have been presented. It is hoped that this paper can serve as a guide for MWI researchers and practitioners, especially those new to the field to comprehend the potential of MWI as a viable imaging tool for breast imaging.

This paper describes a U-net based Deep Learning (DL) approach in combination with Subspace-Based Variational Born Iterative Method (SVBIM) to provide a solution for the quantitative reconstruction of scatterer from the measured scattered field. The proposed technique can be used as an alternative to conventional time consuming and computationally complex iterative methods. This technique comprises a numerical solver (SVBIM) for generating the initial contrast function and a DL network to reconstruct the scatterer profile from the initial contrast function. Further, the proposed technique is validated against theoretical and experimental results available from the literature. Root Mean Square Error (RMSE) value is used as the metric to measure the accuracy of the reconstructed image. The RMSE values of the proposed method show a significant reduction in the reconstruction error compared with the recent Back Propagation-Direct Sampling Method (BP-DSM). The proposed method produces an RMSE value of 0.0813 against 0.1070 in the case of simulation (Austria Profile). The error value obtained by validating against the FoamDielExt experimental database in the case of the proposed method is 0.1037 against 0.1631 reported for BP-DSM method.

No abstract

A deep learning-based approach in conjugation with Fourier Diffraction Theorem (FDT) is proposed in this paper to solve the inverse scattering problem arising in microwave imaging. The proposed methodology is adept in generating a permittivity mapping of the object in less than a second and hence has the potential for real-time imaging. The reconstruction of the dielectric permittivity from the measured scattered field values is done in a single step as against that by a long iterative procedure employed by conventional numerical methods. The proposed technique proceeds in two stages; with the initial estimate of the contrast function being generated by the FDT in the first stage. This initial profile is fed to a trained U-net to reconstruct the final dielectric permittivities of the scatterer in the second stage. The capability of the proposed method is compared with other works in the recent literature using the Root Mean Square Error (RMSE). The proposed method generates an RMSE of 0.0672 in comparison to similar deep learning methods like Back Propagation-Direct Sampling Method (BP-DSM) and Subspace-Based Variational Born Iterative Method (SVBIM), which produce error values 0.1070 and 0.0813 in the case of simulation (using Austria Profile). The RMSE level while reconstructing the experimental data (FoamDielExt experimental database) is 0.0922 for the proposed method as against 0.1631 and 0.1037 for BP-DSM and SVBIM, respectively.

This paper describes an innovative technique for the quantitative reconstruction of the dielectric and conductivity distribution of objects in a microwave tomography framework using sparse data. The proposed method tries to extract information about the size, shape, localisation, and dielectric distribution of various inclusions within the object under study using an iterative reconstruction methodology in the sparse domain. The proposed algorithm combines the Distorted Born Iteration method (DBIM) and a convex optimization technique for solving the inverse ill-posed problem. The Re-weighted Basis Pursuit (RwBP) algorithm is chosen as the convex optimization technique in this work. The performance of the proposed algorithm has been compared with the TV-norm method, and the results obtained are highly encouraging. The proposed method produces a significant reduction in the reconstruction error as compared to the TV norm method with an error value of 0.083 as against 0.32 in the case of TV norm in the presence of 25 dB noise. By accurately preserving the edges of the inclusions, the proposed technique is found to provide an overall improvement in the reconstruction in terms of tissue differentiation (permittivity and conductivity), dimensions of inclusions, resolution, shape, size, and coordinate localisation of inclusions. The proposed algorithm converges within 10-12 iterations as compared to other complex imaging algorithms available in the literature. Further, this proposed technique is validated using experimental data from an actual breast imaging setup. The three inclusions of 10 mm, 6 mm, and 3 mm have been localised with errors of 0.052, 0.04, and 0.09, respectively. The results obtained from the real-time data show the applicability and feasibility of the proposed algorithm in breast tumor imaging application.