Ilaria Catapano

A FEASIBILITY STUDY ON MICROWAVE IMAGING FOR BRAIN STROKE MONITORING

The adoption of microwave imaging as a tool for non- invasive monitoring of brain stroke has recently gained increasing attention. In this respect, the paper aims at providing a twofold contribution. First, we introduce a simple design tool to devise guidelines to properly set the working frequency as well as to choose the optimum matching medium needed to facilitate the penetration of the probing wave into the head. Second, we propose an imaging strategy based on a modifled formulation of the linear sampling method, which allows a quasi real time monitoring of the diseases evolution. The accuracy of the design guidelines and performance of the imaging strategy are assessed through numerical examples dealing with 2D anthropomorphic phantoms.

Microwave Cancer Imaging Exploiting Magnetic Nanoparticles as Contrast Agent

In this paper, a microwave technique for breast cancer imaging is presented. The approach is based on the use of magnetic nanoparticles as contrast agent to induce a nonnull magnetic contrast selectively localized within the tumor. This allows us to face cancer imaging as the reconstruction of a magnetic contrast from the corresponding scattered field. To extract, from the measured data the contribution due to the magnetic contrast, i.e., the signal meaningful for cancer imaging, the approach exploits the possibility of modulating the magnetic response of magnetic nanoparticles by means of a polarizing magnetic field. The achievable reconstruction capabilities and the robustness against uncertainties on the electric features of the surrounding electric scenario are assessed by means of numerical examples.

Improved Sampling Methods for Shape Reconstruction of 3-D Buried Targets

This paper addresses the problem of reconstructing geometrical features of 3-D targets embedded into a nonaccessible region from multiview multistatic scattered field data. Sampling methods (SM) are simple and computationally effective approaches to pursue this task. However, their implementation requires a large number of multipolarization sources and probes. Moreover, their performances are often unsatisfactory for aspect-limited measurement configurations and lossy media. In order to tackle these drawbacks, usually faced in subsurface imaging, we propose a simplified and improved formulation based on the physical interpretation of SM. In particular, such a formulation relies on a small number of single polarization probes and exploits multifrequency data, for the first time in the framework of SM. The performances of the resulting approach are verified by monitoring 3-D regions of large extent.

On Quantitative Microwave Tomography of Female Breast

Microwave tomography deserves attention in biomedical imaging, owing to its potential capability of providing a morphological and functional assessment of the inspected tissues. However, such a goal requires the not trivial task of solving a non linear inverse scattering problem. In this paper, the factors afiecting the complexity of the inverse problem are exploited to trace guidelines aimed at setting the matching ∞uid, the frequency range and the number of probes in such a way that the dielectric parameters of female breast tissues can be reliably retrieved. Examples, concerning 2D realistic numerical phantoms obtained by NMR images, are given to asses a posteriori the efiectiveness of the proposed guidelines.

The Linear Sampling Method as a Way to Quantitative Inverse Scattering

The linear sampling method (LSM) is a simple and effective approach to image the shape of unknown targets via the solution of a linear inverse problem. In this paper, we show that the LSM can also be exploited to devise a novel effective approximation of the scattering phenomenon, that leads to a new noniterative linear inversion method for the estimation of the targets electric contrast. Since the introduced approximation relies on the broad applicability of the LSM, the proposed inversion method is suitable to tackle inverse scattering problems involving nonweak scatterers. As such, it represents an innovative, yet effective, way to tackle quantitative imaging. Examples with numerical and experimental data are given to show the performance of the approach. In particular, results obtained with Fresnel data-sets show that the proposed method is capable of successfully imaging targets which have been so far processed using nonlinear iterative schemes and taking advantage of frequency diversity.

An Effective Procedure for MNP-Enhanced Breast Cancer Microwave Imaging

Magnetic nanoparticles-enhanced microwave imaging has been recently proposed to overcome the limitations of conventional microwave imaging methods for breast cancer monitoring. In this paper, we discuss how to tackle the linear inverse scattering problem underlying this novel technique in an effective way. In particular, our aim is to minimize the required a priori patient-specific information, avoid occurrence of false positives, and keep the computational burden low. By relying on an extensive numerical analysis in realistic conditions, we show that the method can provide accurate and reliable images without information on the inner structure of the inspected breast and with an only rough knowledge of its shape. Notably, this allows moving to an offline stage the computationally intensive part of the image formation procedure. In addition, we show how to appraise the total amount of magnetic contrast agent targeted in the tumor.

Optimal Constrained Field Focusing for Hyperthermia Cancer Therapy: a Feasibility Assessment on Realistic Phantoms

Microwave hyperthermia is a non-invasive treatment for cancer which exploits a selective heating of tissues induced through focused electromagnetic flelds. In order to improve the treatments e-ciency, while minimizing side efiects, it is necessary to achieve a constrained focusing of the fleld radiated by the sources. To address this issue, in this paper we present an innovative and computationally efiective approach to the fleld focusing for hyperthermia. The proposed method, after establishing the number of sources to be used, determines the excitations of the given set of sources such to produce a maximum fleld in a given region of space, subject to a completely arbitrary mask for the fleld amplitude in all other regions. As the approach relies on a formulation of the problem in terms of convex programming, it is able to achieve the globally optimal solution without the adoption of computationally intensive global optimization procedures. A preliminary assessment of feasibility is given on hyperthermia therapy of breast cancer by means of numerical examples run on realistic 2D phantoms of female breast.

An Imaging Method for Concealed Targets

A novel microwave imaging method to detect and possibly characterize nonaccessible targets concealed into a wall or a floor from the measured backscattered field is proposed. The problem is cast as an inverse scattering one aimed at imaging a piecewise homogeneous target embedded in a layered structure. The proposed method is based on a linear sampling method (LSM), a general and computationally effective shape-reconstruction approach that determines the targets morphology through an indicator defined as the regularized solution of a linear inverse problem. However, LSM cannot discern between homogeneous and inhomogeneous targets; hence, it may retrieve the cache but not characterize its content. To overcome this drawback, we propose a two-step approach in which the cache is first imaged by using the standard LSM indicator. Then, by taking advantage of the physical interpretation of the LSM and of the first steps result, a modified indicator is defined and exploited to detect the inclusions possibly hidden in the retrieved cache. Numerical examples are provided as a proof of concept.

SAR Imaging Algorithms and Some Unconventional Applications: A unified mathematical overview

This article deals with two significant aspects related to synthetic aperture radar imaging (SAR-I) of relevant theoretical and applicative interest. The first objective regards the analysis of the most-used SAR-I approaches under the unified mathematical framework provided by the Porter-Bojarski integral equation. The second objective is to provide an updated overview on how SAR-I research is generalizing previous algorithms to deal with unconventional scenarios.

Wavelet-Based Adaptive Multiresolution Inversion for Quantitative Microwave Imaging of Breast Tissues

Microwave imaging is a feasible tool for medical imaging owing to the different electric properties of tissues. To this end, it is important to devise inversion procedures capable of achieving a reliable quantitative estimate of the tissues under test. With respect to breast screening, we propose a full-wave inversion method, which takes advantage of the adaptive multiresolution features of the wavelet basis to accommodate the trade off between spatial resolution and inversion stability. As shown through a preliminary numerical assessment on 2D anthropomorphic breast phantoms, the proposed approach is capable of achieving satisfying results without any a priori knowledge of breast shape, as well as of thickness and electric properties of the skin layer.