Search

Magnetic Resonance imaging based Electrical Properties Tomography (MR-EPT) is a non-invasive technique that measures the electrical properties (EPs) of biological tissues. In this work, we present and numerically investigate the performance of an unrolled, physics-assisted method for 2D MR-EPT reconstructions, where a cascade of Convolutional Neural Networks is used to compute the contrast update. Each network takes in input the EPs and the gradient descent direction (encoding the physics underlying the adopted scattering model) and returns as output the updated contrast function. The network is trained and tested in silico using 2D slices of realistic brain models at 128 MHz. Results show the capability of the proposed procedure to reconstruct EPs maps with quality comparable to that of the popular Contrast Source Inversion-EPT, while significantly reducing the computational time.

The canonical problem of designing the complex excitations feeding an array such to ensure a desired field intensity distribution inside a given region of interest, while keeping it under control in some other regions, is addressed. To this end, an auxiliary physics inspired model for the induced total field is proposed, whose off-line analysis allows a simplified approach to understand the convenient and non-convenient field interferences between canonical solutions for the induced total field, that are zero order Bessel functions. Moreover, its analysis drastically reduces the computational burden associated to the multi-control points based approaches. The problem at hand plays a key role in many different applications, including radio communications, wireless power transfer as well as hyperthermia treatment planning, and in this paper attention is paid to radiofrequency shimming in magnetic resonance imaging. In fact, the proposed model and tools are tailored to the challenging case of leveling of the magnetic field intensity within an MRI scanner and in case of a bidimensional realistic head phantom.

Quantitative microwave imaging employing an ideal Veselago lens is demonstrated using synthetic 2D experiments. Reconstructions of the complex-valued dielectric constant of an arbitrary object-of-interest are obtained from noiseless data using a non-iterative technique. A closed-form Veselago lens Green’s function is utilized in the formulation of the problem. The contrast sources corresponding to a single illumination of the object-of-interest is recovered using total-field measurements at several receiver points located in the lens’ focusing region. Collecting the field in the presence of the ideal Veselago lens produces well-conditioned matrices for the discretized inverse source problem. The contrast is then directly obtained by first calculating the total-field within the object. Synthetic examples having features as small as $\lambda /25$ show that the resolution is only limited by the signal-to-noise ratio of data. Experiments with noisy data reveals that the inversion procedure directly mirrors the noise in the data. Two new filtering processes are introduced based on the truncated Singular Value Decomposition of the data operator and the domain operator involved in the inversion process. These successfully deal with noisy reconstructions. Results show that using the new filtering methods, even with a single illuminating source, outperforms the least-squares method with several illuminating sources.

At the local scale, environmental parameters often require monitoring by means of affordable measuring techniques and technologies given they need to be frequently surveyed. Streamflow in riverbeds or in channels is a hydrological variable that needs to be monitored in order to keep the runoff regimes under control and somehow forecast floods, allowing prevention of damage for people and infrastructure. Moreover, measuring such a variable is always extremely important for the knowledge of the environmental status of connected aquatic ecosystems. This paper presents a new approach to assessing hydrodynamic features related to a given channel by means of a beamforming technique that was applied to video sensing. Different features have been estimated, namely the flow velocity, the temperature, and the riverbed movements. The applied beamforming technique works on a modified sum and delay method, also using the Multiple Signal Classification algorithm (MUSIC), by acting as Synthetic Aperture Radar (SAR) post-processing. The results are very interesting, especially compared to the on-site measured data and encourage the use of affordable video sensors located along the channel or river course for monitoring purposes. The paper also illustrates the use of beamforming measurements to be calibrated by means of conventional techniques with more accurate data. Certainly, the results can be improved; however, they indicate some margins of improvements and updates. As metrics of assessment, a histogram of greyscale/pixels was adopted, taking into account the example of layers and curve plots. They show changes according to the locations where the supporting videos were obtained.

Theoretical predictions as well as experiments performed at storage rings have shown that the lifetimes of β-radionuclides can change significantly as a function of the ionization state. In this paper we describe an innovative approach, based on the use of a compact plasma trap to emulate selected stellar-like conditions. It has been proposed within the PANDORA project (Plasmas for Astrophysics, Nuclear Decay Observation and Radiation for Archaeometry) with the aim to measure, for the first time in plasma, nuclear β-decay rates of radionuclides involved in nuclear-astrophysics processes. To achieve this task, a compact magnetic plasma trap has been designed to reach the needed plasma densities, temperatures, and charge-states distributions. A multi-diagnostic setup will monitor, on-line, the plasma parameters, which will be correlated with the decay rate of the radionuclides. The latter will be measured through the detection of the γ-rays emitted by the excited daughter nuclei following the β-decay. An array of 14 HPGe detectors placed around the trap will be used to detect the emitted γ-rays. For the first experimental campaign three isotopes, 176Lu, 134Cs, and 94Nb, were selected as possible physics cases. The newly designed plasma trap will also represent a tool of choice to measure the plasma opacities in a broad spectrum of plasma conditions, experimentally poorly known but that have a great impact on the energy transport and spectroscopic observations of many astrophysical objects. Status and perspectives of the project will be highlighted in the paper.