Breast cancer represents one of the most impactful diseases for both human health and healthcare system. The use of cutting-edge and innovative technology solutions is paramount to allow a proper and early diagnosis, which also would mitigate its impact on the society. In this framework, the proposed work explore the use of safe, non-ioninzing microwave radiation with dense deep learning models for the detection and localisation of breast tumor, providing spatial information to support medical decision and personalised medicine. After a screening step for the identification of malignant breasts, a tumor spatial probability map is estimated by means of fully -connected deep learning models. To this aim, an in-house, two-dimensional MRI-derived breast database consisting of 160,000 profiles was adopted, and the localisation performance was quantified by adopting proper quality metrics. The obtained results confirm the potentialities that deep learning and microwave technology can have as emerging solutions in the healthcare system for improving the quality of patients' life.

Magnetoencephalography represents one of the main state-of-the-art methods for brain functional analysis. It is a non-invasive device where a helmet equipped with sensors is located around the scalp in order to measure magnetic fields originated from the electrical brain activity. In many studies magnetoencephalography signals are exploited to map the brain function in order to identify sources locations and reconstruct their temporal signals. This is possible through the resolution of an inverse problem that passes from the recorded signals to the brain activity. Several reconstruction algorithms have been proposed and exploited in literature but, due to the inherent ill-posedness of the inverse problem, they offer good temporal resolution but a limited spatial resolution. Many clinical applications require precise localization of pathological tissues to provide reliable information for neurologists. Within this manuscript, an alternative inversion algorithm is proposed in order to estimate a more refined identification of source locations. The proposed one is a deep learning algorithm named deep-MEG. It demonstrates to overcome other state-of-the-art reconstruction approaches being effective in precisely locating the active brain area and in reconstructing its temporal evolution with a good level of accuracy, even in a challenging noisy scenario. Moreover, deep-Megcan reconstruct active areas belonging to the whole brain and it is not limited to cortical sources.

This paper introduces a new system for non-contact hand movement sensing, utilizing ultrasound sensors with millimeter precision. Acting akin to a sonar, the system detects pressure waves reflected from the moving hand. Due to the high spatio/temporal resolution of the system, the proposed solution has been tested for finger tapping monitoring. Finger tapping serves as a key clinical test for diagnosing neurodegenerative diseases, such as Parkinson's disease, where symptoms like bradykinesia, rigidity, and tremor are focal points of investigation. Currently, clinicians visually assess bradykinesia by observing subjects' finger tapping gestures, rating the severity qualitatively. The idea is to provide a quantitative assessment of the tapping providing a support for diagnosis. In detail, through subsequent time/frequency analysis and specialized processing, tapping peaks are identified, enabling evaluation of tapping stability by measuring interpeak distances. This feasibility study presents the prototype and initial measurement findings.

The accurate quantitative estimation of the electromagnetic properties of tissues can serve important diagnostic and therapeutic medical purposes. Quantitative microwave tomography is an imaging modality that can provide maps of the in-vivo electromagnetic properties of the imaged tissues, i.e. both the permittivity and the electric conductivity. A multi-step microwave tomography approach is proposed for the accurate retrieval of such spatial maps of biological tissues. The underlying idea behind the new imaging approach is to progressively add details to the maps in a step-wise fashion starting from single-frequency qualitative reconstructions. Multi-frequency microwave data is utilized strategically in the final stage. The approach results in improved accuracy of the reconstructions compared to inversion of the data in a single step. As a case study, the proposed workflow was tested on an experimental microwave data set collected for the imaging of the human forearm. The human forearm is a good test case as it contains several soft tissues as well as bone, exhibiting a wide range of values for the electrical properties.

Biometrics showed their usefulness in several applications, systems with fingerprint or face identification are well accepted in several fields with possibly different constraints like, for example, forensic or smartphone unlocking. In this paper, a novel ultrasound prototype for person identification and authentication has been designed, built and tested. The system, which acts as a sonar, measures the pressure wave backscattered from a moving hand. Such signal is subsequently processed by means of time/frequency analysis and a deep learning detector is implemented in order to identify/authenticate the user based on some peculiarity in the gestures execution. The proposed solution is cheap and, due to the high adaptability, allows different security levels. Promising results are obtained after tests carried out with the help of 10 volunteers.

Subsurface non-destructive exploration is of interest for several applications which span from archeology and civil engineering to safety and security. Among the available remote sensing imaging modalities, ground penetrating radar (GPR) seems very attractive for the detection and characterisation of buried targets. This technology exploits electromagnetic signals in the microwave band to perform the exploration in a non-invasive way. Unfortunately, conventional GPR images require an expert user to be interpreted and do not provide quantitative information about the buried objets. In this framework, this paper explores the use of artificial neural networks for quantitative multiple-input-multiple-output tomographic ground penetrating radar to improve the reliability of the quantitative tomographic recovery, speeding up the nonlinear inversion and providing a user-friendly image which is easy to be interpreted even for a non-expert user.

In this paper, an experimental assessment of the linear sampling method (LSM) on the Georgia Institute of Technology multi-frequency multistatic dataset is proposed. The employed approach belongs to the class of linear, qualitative approaches, which seems to be quite promising due to the low computational complexity. Such a method only involves a singular value decomposition (SVD) of the scattering matrix, and this feature, together with its ease of implementation, make it a good candidate for the use in several practical applications.

Automatic Target Recognition (ATR) aims at detecting the presence and at recognizing the typology and the orientation of targets within a scenario, by using an unsupervised approach. In Syntethic Aperture Radar imaging this turns to be a difficult task due to the specific characteristics of clutter and background noise. Within this manuscript a new two-steps ATR algorithm based on Kolmogorov-Smirnov test is presented. The method has been tested on real MSTAR datasets showing interesting performances.

Ground penetrating radar (GPR) represents a promising technology for the non-invasive exploration of soil and for the quantitative characterization and localization of buried objects. Unfortunately, this technique suffers for some limitations, among which the high number of data required for the processing of GPR information. In order to overcome this drawback, the paper proposes a mixed-norm approach based on the combination of compressive sensing (CS) theory and Wavelet decomposition basis to enhance signal processing and meantime to reduce the number of data required for the inversion procedure. The accuracy of the proposed approach is validated by means of a numerical analysis carried out in simplified two-dimensional (2D) scenarios.

Microwave imaging represent an emerging tool in the framework of noninvasive diagnostics, due to the potential advantages of providing quantitative characterizations of the inspected domains. In general, handling such a problem is not an easy task, but under some simplifying hypotheses it is possible to reduce the ill-posedness of the inverse problem and even to employ some linearization strategies. In this paper, a robust method for the quantitative imaging of weak buried objects in the framework of Ground Penetrating Radar applications by exploiting the theory of Compressed Sensing (CS) and Wavelet decomposition (WT) in a canonical two-dimensional configuration is proposed.

Microwave imaging (MWI) techniques represent an assessed tool to face a lot of diagnostic problems in which a non-invasive analysis of an unaccessible area is required, since they allow to obtain qualitative and quantitative information about the nature of the targets embedded inside an imaging domain. In this communication, the authors propose an application of MWI techniques to a well-known aspect-limited configuration, that is the one proposed in Ground Penetrating Radar for demining application. More in details, we applied the theory of Compressive Sampling (CS) in order to improve the quality of recoveries and, in the meanwhile, reduce the number of data acquired and employed in a 3-D fashion in the case of low-moderate dielectric objects, and therefore the well-known Born Approximation has been exploited in the inversion algorithm.

It is well-known that in a canonical inverse scattering problem there is only a limited amount of independent data which is quantified by the degree of freedom of the problem one is dealing with. Of course, it is mandatory, in order to recover the signal in a proper way, to sample the data at least at Nyquist rate. However, there are some cases in which the number of independent data of a signal is much smaller than what its bandwidth seems to suggest, and in these situations the theory of Compressive Sensing may help to improve reconstruction capabilities without using so much information. In this framework, the following communication deals with a preliminary analysis of the solution of microwave imaging problems by exploiting the theory of Compressive Sensing in a linearized multiview-multistatic single-frequency approach.