Search

This paper proposes an innovative technique to calibrate microwave imaging (MWI) systems combining available measured data with simulated synthetic ones. The introduced technique aims to compensate the variations of the antenna array due to unavoidable manufacturing tolerances and placement, in comparison to the nominal electromagnetic (EM) scenario. The scheme is tested virtually and experimentally for the MWI of the adult human head tissues. The virtual EM analysis uses a realistic 3-D CAD model working together with a full-wave software, based on the finite element method. Meanwhile, the real implementation employs a single-cavity anthropomorphic head phantom and a custom brick-shaped antenna array working at around 1 GHz.

Brain stroke is one of the most common diseases worldwide. Magnetic resonance imaging and computerized tomography are currently the two most assessed and reliable diagnostic tools, of great support to the clinicians. However, these devices are costly, time consuming and not portable, so that they are not appropriate for bedside monitoring. Hence, there is an increasing worldwide interest to develop complementary techniques to continuously monitor a patient after the stroke onset. In this paper, we report about the experimental testing of a novel device able to monitor cerebrovascular diseases, whose realization is based on microwave imaging technology. A 2-D version of the device has been fully experimentally tested and its 3-D version is under development.

The present paper describes our on-going work towards the realization of a novel device to perform continuous monitoring of a patient after a stroke onset. The realization of the antenna array and of the control electronics will be discussed, together with a first experimental test of the system.

This work presents some preliminary results from applying our previously developed microwave imaging algorithm based on two-step iterative shrinkage/thresholding (TwIST) to data measured from an experimental microwave imaging system. Combining the distorted Born iterative method (DBIM) with the TwIST linear solver, our two-dimensional (2-D) algorithm is applied to reconstruct 2-D slices of the complex permittivity of the interrogated imaging domain. Experimental data are obtained by rotating a two-antenna system along the cylindrical imaging domain, which is filled with Triton X-100. The imaging object is an anatomically realistic breast phantom with a tube filled water representing the tumor-like target. Measurements are covered in the frequency range from 0.5 to 4.0 GHz. The paper presents reconstructed images at 900 MHz, while more results will be presented at the conference.

In this paper we describe the experimental testing of our microwave imaging (MWI) system prototype for use in research towards early breast cancer detection. The main characteristics of the proposed MWI prototype are the low cost and the portability due to the use, for signals generation and acquisition, of low-cost yet high-precision off-the-shelf components (COTS) that replace a costly vector network analyzer. Good detections are reported for simple breast phantoms.

In this paper we presents experimental results of a designed microwave-imaging system prototype combined with the interferometric multiple signal classification (I-MUSIC) algorithm for use in research towards early breast cancer detection.

International audience; This paper presents the design of the layout of the radiating part of a microwave imaging system for brain stroke monitoring. The radiating part of the system is constituted by an array of antennas placed tangent to a surface conformal to the upper part of head and the design procedure addresses the positioning and the number of the array elements. To this end, an analysis of the relevant scattering operator is performed, taking into account the limited dynamic of the measurement system and the limited number of implemented antennas. Finally, a preliminary tomographic image of a hemorrhagic stroke obtained with synthetic data is reported.