In-Vivo Electrical Properties Estimation of Biological Tissues by Means of a Multi-Step Microwave Tomography Approach.

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.

Microwave Breast Sensing via Deep Learning for Tumor Spatial Localization by Probability Maps.

BACKGROUND: microwave imaging (MWI) has emerged as a promising modality for breast cancer screening, offering cost-effective, rapid, safe and comfortable exams. However, the practical application of MWI for tumor detection and localization is hampered by its inherent low resolution and low detection capability. METHODS: this study aims to generate an accurate tumor probability map directly from the scattering matrix. This direct conversion makes the probability map independent of specific image formation techniques and thus potentially complementary to any image formation technique. An approach based on a convolutional neural network (CNN) is used to convert the scattering matrix into a tumor probability map. The proposed deep learning model is trained using a large realistic numerical dataset of two-dimensional (2D) breast slices. The performance of the model is assessed through visual inspection and quantitative measures to assess the predictive quality at various levels of detail. RESULTS: the results demonstrate a remarkably high accuracy (0.9995) in classifying profiles as healthy or diseased, and exhibit the model's ability to accurately locate the core of a single tumor (within 0.9 cm for most cases). CONCLUSION: overall, this research demonstrates that an approach based on neural networks (NN) for direct conversion from scattering matrices to tumor probability maps holds promise in advancing state-of-the-art tumor detection algorithms in the MWI domain.

A Deep Learning Approach for Diagnosis Support in Breast Cancer Microwave Tomography.

In this paper, a deep learning technique for tumor detection in a microwave tomography framework is proposed. Providing an easy and effective imaging technique for breast cancer detection is one of the main focuses for biomedical researchers. Recently, microwave tomography gained a great attention due to its ability to reconstruct the electric properties maps of the inner breast tissues, exploiting nonionizing radiations. A major drawback of tomographic approaches is related to the inversion algorithms, since the problem at hand is nonlinear and ill-posed. In recent decades, numerous studies focused on image reconstruction techniques, in same cases exploiting deep learning. In this study, deep learning is exploited to provide information about the presence of tumors based on tomographic measures. The proposed approach has been tested with a simulated database showing interesting performances, in particular for scenarios where the tumor mass is particularly small. In these cases, conventional reconstruction techniques fail in identifying the presence of suspicious tissues, while our approach correctly identifies these profiles as potentially pathological. Therefore, the proposed method can be exploited for early diagnosis purposes, where the mass to be detected can be particularly small.

An End-to-End Deep Learning Approach for Quantitative Microwave Breast Imaging in Real-Time Applications.

(1) Background: In this paper, an artificial neural network approach for effective and real-time quantitative microwave breast imaging is proposed. It proposes some numerical analyses for the optimization of the network architecture and the improvement of recovery performance and processing time in the microwave breast imaging framework, which represents a fundamental preliminary step for future diagnostic applications. (2) Methods: The methodological analysis of the proposed approach is based on two main aspects: firstly, the definition and generation of a proper database adopted for the training of the neural networks and, secondly, the design and analysis of different neural network architectures. (3) Results: The methodology was tested in noisy numerical scenarios with different values of SNR showing good robustness against noise. The results seem very promising in comparison with conventional nonlinear inverse scattering approaches from a qualitative as well as a quantitative point of view. (4) Conclusion: The use of quantitative microwave imaging and neural networks can represent a valid alternative to (or completion of) modern conventional medical imaging techniques since it is cheaper, safer, fast, and quantitative, thus suitable to assist medical decisions.

An Experimental Ultrasound System for Qualitative Tomographic Imaging.

The advancement of new promising techniques in the field of biomedical imaging has always been paramount for the research community. Recently, ultrasound tomography has proved to be a good candidate for non-invasive and safe diagnostics. In particular, breast cancer imaging may benefit from this approach, as frequent screening and early diagnosis require decreased system size and costs compared to conventional imaging techniques. Furthermore, a major advantage of these approaches consists in the operator-independent feature, which is very desirable compared to conventional hand-held ultrasound imaging. In this framework, the authors present some imaging results on an experimental campaign acquired via an in-house ultrasound tomographic system designed and built at the University of Naples Parthenope. Imaging performance was evaluated via different tests, showing good potentiality in structural information retrieval. This study represents a first proof of concept in order to validate the system and to consider further realistic cases in near future applications.

Hand Gesture Signatures Acquisition and Processing by Means of a Novel Ultrasound System.

Hand gestures represent a natural way to express concepts and emotions which are peculiar to each culture. Several studies exploit biometric traits, such as fingerprint, iris or face for subject identification purposes. Within this paper, a novel ultrasound system for person identification that exploits hand gestures is presented. The system works as a sonar, measuring the ultrasonic pressure waves scattered by the subject's hand, and analysing its Doppler information. Further, several transformations for obtaining time/frequency representations of the acquired signal are computed and a deep learning detector is implemented. The proposed system is cheap, reliable, contactless and can be easily integrated with other personal identification approaches allowing different security levels. The performances are evaluated via experimental tests carried out on a group of 25 volunteers. Results are encouraging, showing the promising potential of the system.

A Multithreshold Iterative DBIM-Based Algorithm for the Imaging of Heterogeneous Breast Tissues.

OBJECTIVE: This paper proposes a novel microwave imaging (MWI) multifrequency technique, which combines compressive sensing (CS) with the well-known distorted Born iterative method. CS strategies are emerging as a promising tool in MWI applications, which can improve reconstruction quality and/or reduce the number of data samples. METHODS: The proposed approach is based on iterative shrinkage thresholding algorithm (ISTA), which has been modified to include an automatic and adaptive selection of multithreshold values. RESULTS: This adaptive multithreshold ISTA implementation is applied in reconstruction of two-dimensional (2-D) numerical heterogeneous breast phantoms, where it outperforms the standard thresholding implementation. We show that our approach is also successful in 3-D simulations of a realistic imaging experiment, despite the mismatch between the data and our algorithm's forward model. CONCLUSION: These results suggest that the proposed algorithm is a promising tool for medical MWI applications. SIGNIFICANCE: Important novelties of this approach are the use of multiple thresholds to recover the different unknowns in the Debye model as well as the adaptive selection of these thresholds. Moreover, we have shown that employing modified hard constraints inside the linear step of the inversion procedure can enhance reconstruction quality.

Denoising of MR images using Kolmogorov-Smirnov distance in a Non Local framework.

Data coming from any acquisition system, such as Magnetic Resonance Imaging ones, are affected by noise. Although modern high field scanners can reach high Signal to Noise Ratios, in some circumstances, for example in case of very weak signals due to a specific acquisition sequence, noise becomes a critical issue that has to be properly handled. In the last years methods based on the so called Non Local Mean have proven to be very effective in denoising tasks. The idea of these filters is to find similar patches across the image and to jointly exploit them to obtain the restored image. A critical point is the distance metric adopted for measuring similarity. Within this manuscript, we propose a filtering technique based on the Kolmogorov-Smirnov distance. The main innovative aspect of the proposed method consists of the criteria adopted for finding similar pixels across the image: it is based on the statistics of the points rather than the widely adopted weighted Euclidean distance. More in details, the Cumulative Distribution Functions of different pixels are evaluated and compared in order to measure their similarities, exploiting a stack of images of the same slice acquired with different acquisition parameters. To quantitatively and qualitatively assess the performances of the approach, a comparison with other widely adopted denoising filters in case of both simulated and real datasets has been carried out. The obtained results confirm the validity of the proposed solution.

Compressive Sensing for Breast Microwave Imaging.

This work proposes a novel microwave imaging (MWI) multi-frequency technique, which combines compressive sensing (CS) with the well-known distorted Born iterative method (DBIM) to enhance the accuracy in the imaging procedure. CS strategies are emerging as a promising tool in MWI applications, which can also reduce the number of data samples. The proposed approach is based on an iterative shrinkage thresholding algorithm (ISTA), which has been modified to include an automatic and adaptive selection of multi-threshold values. The proposed implementation is applied in reconstruction of two-dimensional numerical heterogeneous breast phantoms, where it outer-performs the standard thresholding implementation and proves to be an interesting tool for medical imaging applications.

Enhanced Wiener filter for ultrasound image restoration.

BACKGROUND AND OBJECTIVE: Speckle phenomenon strongly affects UltraSound (US) images. In the last years, several efforts have been done in order to provide an effective denoising methodology. Although good results have been achieved in terms of noise reduction effectiveness, most of the proposed approaches are not characterized by low computational burden and require the supervision of an external operator for tuning the input parameters. METHODS: Within this manuscript, a novel approach is investigated, based on Wiener filter. Working in the frequency domain, it is characterized by high computational efficiency. With respect to classical Wiener filter, the proposed Enhanced Wiener filter is able to locally adapt itself by tuning its kernel in order to combine edges and details preservation with effective noise reduction. This characteristic is achieved by implementing a Local Gaussian Markov Random Field for modeling the image. Due to its intrinsic characteristics, the computational burden of the algorithm is sensibly low compared to other widely adopted filters and the parameter tuning effort is minimal, being well suited for quasi real time applications. RESULTS: The approach has been tested on both simulated and real datasets, showing interesting performances compared to other state of art methods. CONCLUSIONS: A novel denoising method for UltraSound images is proposed. The approach is able to combine low computational burden with interesting denoising performances and details preservation.

A 3D MRI denoising algorithm based on Bayesian theory.

BACKGROUND: Within this manuscript a noise filtering technique for magnetic resonance image stack is presented. Magnetic resonance images are usually affected by artifacts and noise due to several reasons. Several denoising approaches have been proposed in literature, with different trade-off between computational complexity, regularization and noise reduction. Most of them is supervised, i.e. requires the set up of several parameters. A completely unsupervised approach could have a positive impact on the community. RESULTS: The method exploits Markov random fields in order to implement a 3D maximum a posteriori estimator of the image. Due to the local nature of the considered model, the algorithm is able do adapt the smoothing intensity to the local characteristics of the images by analyzing the 3D neighborhood of each voxel. The effect is a combination of details preservation and noise reduction. The algorithm has been compared to other widely adopted denoising methodologies in MRI. Both simulated and real datasets have been considered for validation. Real datasets have been acquired at 1.5 and 3 T. The methodology is able to provide interesting results both in terms of noise reduction and edge preservation without any supervision. CONCLUSIONS: A novel method for regularizing 3D MR image stacks is presented. The approach exploits Markov random fields for locally adapt filter intensity. Compared to other widely adopted noise filters, the method has provided interesting results without requiring the tuning of any parameter by the user.

Bayesian MRI denoising in complex domain.

In recent years, several efforts have been done for producing Magnetic Resonance Image scanner with higher magnetic field strength mainly for increasing the Signal to Noise Ratio and the Contrast to Noise Ratio of the acquired images. However, denoising methodologies still play an important role for achieving images neatness. Several denoising algorithms have been presented in literature. Some of them exploit the statistical characteristics of the involved noise, some others project the image in a transformed domain, some others look for geometrical properties of the image. However, the common denominator consists in working in the amplitude domain, i.e. on the gray scale, real valued image. Within this manuscript we propose the idea of performing the noise filtering in the complex domain, i.e. on the real and on the imaginary parts of the acquired images. The advantage of the proposed methodology is that the statistical model of the involved signals is greatly simplified and no approximations are required, together with the full exploitation of the whole acquired signal. More in detail, a Maximum A Posteriori estimator developed for the handling complex data, which adopts Markov Random Fields for modeling the images, is proposed. First results and comparison with other widely adopted denoising filters confirm the validity of the method.

Intra voxel analysis in magnetic resonance imaging.

A technique for analyzing the composition of each voxel, in the magnetic resonance imaging (MRI) framework, is presented. By combining different acquisitions, a novel methodology, called intra voxel analysis (IVA), for the detection of multiple tissues and the estimation of their spin-spin relaxation times is proposed. The methodology exploits the sparse Bayesian learning (SBL) approach in order to solve a highly underdetermined problem imposing the solution sparsity. IVA, developed for spin echo imaging sequence, can be easily extended to any acquisition scheme. For validating the approach, simulated and real data sets are considered. Monte Carlo simulations have been implemented for evaluating the performances of IVA compared to methods existing in literature. Two clinical datasets acquired with a 3T scanner have been considered for validating the approach. With respect to other approaches presented in literature, IVA has proved to be more effective in the voxel composition analysis, in particular in the case of few acquired images. Results are interesting and very promising: IVA is expected to have a remarkable impact on the research community and on the diagnostic field.

A New Methodology for 3D Target Detection in Automotive Radar Applications.

Today there is a growing interest in automotive sensor monitoring systems. One of the main challenges is to make them an effective and valuable aid in dangerous situations, improving transportation safety. The main limitation of visual aid systems is that they do not produce accurate results in critical visibility conditions, such as in presence of rain, fog or smoke. Radar systems can greatly help in overcoming such limitations. In particular, imaging radar is gaining interest in the framework of Driver Assistance Systems (DAS). In this manuscript, a new methodology able to reconstruct the 3D imaged scene and to detect the presence of multiple targets within each line of sight is proposed. The technique is based on the use of Compressive Sensing (CS) theory and produces the estimation of multiple targets for each line of sight, their range distance and their reflectivities. Moreover, a fast approach for 2D focus based on the FFT algorithm is proposed. After the description of the proposed methodology, different simulated case studies are reported in order to evaluate the performances of the proposed approach.

A Bayesian approach for relaxation times estimation in MRI.

Relaxation time estimation in MRI field can be helpful in clinical diagnosis. In particular, T1 and T2 changes can be related to tissues modification, being an effective tool for detecting the presence of several pathologies and measure their development, thus their estimation is a useful research field. Currently, most techniques work pixel-wise, and transfer the noise reduction task to post processing filters. A novel method for estimating spin-spin and spin-lattice relaxation times is proposed. The approach exploits Markov Random Field theory for modeling the unknown data and implements an a posteriori estimator in the Bayesian framework. The effect is the joint parameters estimation and noise reduction. Proposed methodology, with respect to already existing techniques, is able to provide effective results while preserving details also in case of few acquisitions or severe signal to noise ratio. The algorithm has been tested on both simulated and real datasets.

Speckle reduction based on Wiener filter in ultrasound images.

In this manuscript, a technique for speckle noise reduction in ultrasound images is presented. The method exploits Wiener filter and is able to take into account spatial correlation among noise samples. With respect to classical Wiener filter approach developed in independence hypothesis, the methodology is able to sensibly improve filtering performances, at the cost of no computational time increase. Results on realistic simulated datasets are reported, showing the effectiveness of the approach.

A Novel Statistical Approach for Brain MR Images Segmentation Based on Relaxation Times.

Brain tissue segmentation in Magnetic Resonance Imaging is useful for a wide range of applications. Classical approaches exploit the gray levels image and implement criteria for differentiating regions. Within this paper a novel approach for brain tissue joint segmentation and classification is presented. Starting from the estimation of proton density and relaxation times, we propose a novel method for identifying the optimal decision regions. The approach exploits the statistical distribution of the involved signals in the complex domain. The technique, compared to classical threshold based ones, is able to globally improve the classification rate. The effectiveness of the approach is evaluated on both simulated and real datasets.

Optimal configuration for relaxation times estimation in complex spin echo imaging.

Many pathologies can be identified by evaluating differences raised in the physical parameters of involved tissues. In a Magnetic Resonance Imaging (MRI) framework, spin-lattice T1 and spin-spin T2 relaxation time parameters play a major role in such an identification. In this manuscript, a theoretical study related to the evaluation of the achievable performances in the estimation of relaxation times in MRI is proposed. After a discussion about the considered acquisition model, an analysis on the ideal imaging acquisition parameters in the case of spin echo sequences, i.e., echo and repetition times, is conducted. In particular, the aim of the manuscript consists in providing an empirical rule for optimal imaging parameter identification with respect to the tissues under investigation. Theoretical results are validated on different datasets in order to show the effectiveness of the presented study and of the proposed methodology.

Relaxation time estimation from complex magnetic resonance images.

Magnetic Resonance (MR) imaging techniques are used to measure biophysical properties of tissues. As clinical diagnoses are mainly based on the evaluation of contrast in MR images, relaxation times assume a fundamental role providing a major source of contrast. Moreover, they can give useful information in cancer diagnostic. In this paper we present a statistical technique to estimate relaxation times exploiting complex-valued MR images. Working in the complex domain instead of the amplitude one allows us to consider the data bivariate Gaussian distributed, and thus to implement a simple Least Square (LS) estimator on the available complex data. The proposed estimator results to be simple, accurate and unbiased.

Statistical regularization in linearized microwave imaging through MRF-based MAP estimation: hyperparameter estimation and image computation.

The application of a Markov random fields (MRF) based maximum a posteriori (MAP) estimation method for microwave imaging is presented in this paper. The adopted MRF family is the so-called Gaussian-MRF (GMRF), whose energy function is quadratic. In order to implement the MAP estimation, first, the MRF hyperparameters are estimated by means of the expectation-maximization (EM) algorithm, extended in this case to complex and nonhomogeneous images. Then, it is implemented by minimizing a cost function whose gradient is fully analytically evaluated. Thanks to the quadratic nature of the energy function of the MRF, well posedness and efficiency of the proposed method can be simultaneously guaranteed. Numerical results, also performed on real data, show the good performance of the method, also when compared with conventional techniques like Tikhonov regularization.