Magnetic Resonance Imaging (MRI) is considered today the golden-standard modality for soft tissues. The long acquisition times, however, make it more prone to motion artifacts as well as contribute to the relative high costs of this examination. Over the years, multiple studies concentrated on designing reduced measurement schemes and image reconstruction schemes for MRI, however these problems have been so far addressed separately. On the other hand, recent works in optical computational imaging have demonstrated growing success of simultaneous learning-based design of the acquisition and reconstruction schemes manifesting significant improvement in the reconstruction quality with a constrained time budget. Inspired by these successes, in this work, we propose to learn accelerated MR acquisition schemes (in the form of Cartesian trajectories) jointly with the image reconstruction operator. To this end, we propose an algorithm for training the combined acquisition-reconstruction pipeline end-to-end in a differentiable way. We demonstrate the significance of using the learned Cartesian trajectories at different speed up rates. Code available at https://github.com/tomer196/fastMRI-Cartesian.

The performance of many methods of MRI-based computer-aided diagnosis of breast disease rely on accurate delineation of the breast boundary. This problem has been known to be a challenging one on the account of the complex composition of breast tissue and its extensive inter-subject variability. To address this problem, this paper introduces a new approach to whole-breast segmentation which, as opposed to many existing solutions, can operate in the absence of any prior information on the patient-specific breast anatomy. The proposed algorithm takes advantage of Dijkstra's procedure which allows accurately tracking the boundary between the pectoralis muscle and breast tissue as well as between the breast and its background. The performance of the proposed method has been tested on in vivo MRI volumes and quantified in terms of several performance metrics. The experimental results demonstrate consistent and stable performance of the proposed algorithm in terms of its accuracy and robustness to imaging artefacts.

Surface reconstruction from measurements of spatial gradient is an important computer vision problem with applications in photometric stereo and shape-from-shading. In the case of morphologically complex surfaces observed in the presence of shadowing and transparency artifacts, a relatively large dense gradient measurements may be required for accurate surface reconstruction. Consequently, due to hardware limitations of image acquisition devices, situations are possible in which the available sampling density might not be sufficiently high to allow for recovery of essential surface details. In this paper, the above problem is resolved by means of derivative compressed sensing (DCS). DCS can be viewed as a modification of the classical CS, which is particularly suited for reconstructions involving image/surface gradients. In DCS, a standard CS setting is augmented through incorporation of additional constraints arising from some intrinsic properties of potential vector fields. We demonstrate that using DCS results in reduction in the number of measurements as compared with the standard (dense) sampling, while producing estimates of higher accuracy and smaller variability as compared with CS-based estimates. The results of this study are further supported by a series of numerical experiments.

Surface reconstruction from measurements of spatial gradient is an important computer vision problem with applications in photometric stereo and shape-from-shading. In the case of morphologically complex surfaces observed in the presence of shadowing and transparency artifacts, a relatively large number of gradient measurements may be required for accurate surface reconstruction. Consequently, due to hardware limitations of image acquisition devices, situations are possible in which the available sampling density might not be sufficiently high to allow for recovery of essential surface details. In this paper, the above problem is resolved by means of derivative compressed sensing (DCS). DCS can be viewed as a modification of the classical compressed sensing (CS), which is particularly suited for reconstructions involving image/surface gradients. We demonstrate that using DCS results in substantial data savings as compared to the standard (dense) sampling, while producing estimates of higher accuracy and smaller variability, as compared to CS-base estimates. The results of this study are further supported by a series of numerical experiments.

The problem of reconstruction of digital images from their blurred and noisy measurements is unarguably one of the central problems in imaging sciences. Despite its ill-posed nature, this problem can often be solved in a unique and stable manner, provided appropriate assumptions on the nature of the images to be recovered. In this paper, however, a more challenging setting is considered, in which accurate knowledge of the blurring operator is lacking, thereby transforming the reconstruction problem at hand into a problem of blind deconvolution. As a specific application, the current presentation focuses on reconstruction of short-exposure optical images measured through atmospheric turbulence. The latter is known to give rise to random aberrations in the optical wavefront, which are in turn translated into random variations of the point spread function of the optical system in use. A standard way to track such variations involves using adaptive optics. Thus, for example, the Shack-Hartmann interferometer provides measurements of the optical wavefront through sensing its partial derivatives. In such a case, the accuracy of wavefront reconstruction is proportional to the number of lenslets used by the interferometer and, hence, to its complexity. Accordingly, in this paper, we show how to minimize the above complexity through reducing the number of the lenslets while compensating for undersampling artifacts by means of derivative compressed sensing. Additionally, we provide empirical proof that the above simplification and its associated solution scheme result in image reconstructions, whose quality is comparable to the reconstructions obtained using conventional (dense) measurements of the optical wavefront.

The problem of blind deconvolution of digital images has long been recognized as one of the central problems in imaging science. In this paper, the problem is solved using a hybrid deconvolution approach. Here, the “hybridization” suggests a two-stage reconstruction procedure. In the first stage, some partial information about the point spread function of the imaging system (namely, its magnitude spectrum) is recovered. Subsequently, the obtained information is exploited to explicitly constrain the procedure of inverse filtering. The latter is realized in the form of an optimization problem which is solved using alternating direction method of multipliers (ADMM). We show that this method leads to a particularly efficient numerical scheme, which can be implemented as a succession of analytically computable proximity operations. The effectiveness of the proposed deconvolution procedure is exemplified by a number of computer experiments.

The problem of restoration of digital images from their degraded measurements plays a central role in a multitude of practically important applications. A particularly challenging instance of this problem occurs in the case when the degradation phenomenon is modeled by an ill-conditioned operator. In such a situation, the presence of noise makes it impossible to recover a valuable approximation of the image of interest without using some a priori information about its properties. Such a priori information--commonly referred to as simply priors--is essential for image restoration, rendering it stable and robust to noise. Moreover, using the priors makes the recovered images exhibit some plausible features of their original counterpart. Particularly, if the original image is known to be a piecewise smooth function, one of the standard priors used in this case is defined by the Rudin-Osher-Fatemi model, which results in total variation (TV) based image restoration. The current arsenal of algorithms for TV-based image restoration is vast. In this present paper, a different approach to the solution of the problem is proposed based upon the method of iterative shrinkage (aka iterated thresholding). In the proposed method, the TV-based image restoration is performed through a recursive application of two simple procedures, viz. linear filtering and soft thresholding. Therefore, the method can be identified as belonging to the group of first-order algorithms which are efficient in dealing with images of relatively large sizes. Another valuable feature of the proposed method consists in its working directly with the TV functional, rather then with its smoothed versions. Moreover, the method provides a single solution for both isotropic and anisotropic definitions of the TV functional, thereby establishing a useful connection between the two formulae. Finally, a number of standard examples of image deblurring are demonstrated, in which the proposed method can provide restoration results of superior quality as compared to the case of sparse-wavelet deconvolution.

The problem of reconstruction of digital images from their degraded measurements is regarded as a problem of central importance in various fields of engineering and imaging sciences. In such cases, the degradation is typically caused by the resolution limitations of an imaging device in use and/or by the destructive influence of measurement noise. Specifically, when the noise obeys a Poisson probability law, standard approaches to the problem of image reconstruction are based upon using fixed-point algorithms which follow the methodology first proposed by Richardson and Lucy. The practice of using these methods, however, shows that their convergence properties tend to deteriorate at relatively high noise levels. Accordingly, in the present paper, a novel method for denoising and/or deblurring of digital images corrupted by Poisson noise is introduced. The proposed method is derived under the assumption that the image of interest can be sparsely represented in the domain of a linear transform. Consequently, a shrinkage-based iterative procedure is proposed, which guarantees the solution to converge to the global maximizer of an associated maximum a posteriori criterion. It is shown in a series of computer-simulated experiments that the proposed method outperforms a number of existing alternatives in terms of stability, precision, and computational efficiency.

Despite the relative recency of its inception, the theory of compressive sampling (aka compressed sensing) (CS) has already revolutionized multiple areas of applied sciences, a particularly important instance of which is medical imaging. Specifically, the theory has provided a different perspective on the important problem of optimal sampling in magnetic resonance imaging (MRI), with an ever-increasing body of works reporting stable and accurate reconstruction of MRI scans from the number of spectral measurements which would have been deemed unacceptably small as recently as five years ago. In this paper, the theory of CS is employed to palliate the problem of long acquisition times, which is known to be a major impediment to the clinical application of high angular resolution diffusion imaging (HARDI). Specifically, we demonstrate that a substantial reduction in data acquisition times is possible through minimization of the number of diffusion encoding gradients required for reliable reconstruction of HARDI scans. The success of such a minimization is primarily due to the availability of spherical ridgelet transformation, which excels in sparsifying HARDI signals. What makes the resulting reconstruction procedure even more accurate is a combination of the sparsity constraints in the diffusion domain with additional constraints imposed on the estimated diffusion field in the spatial domain. Accordingly, the present paper describes an original way to combine the diffusion- and spatial-domain constraints to achieve a maximal reduction in the number of diffusion measurements, while sacrificing little in terms of reconstruction accuracy. Finally, details are provided on an efficient numerical scheme which can be used to solve the aforementioned reconstruction problem by means of standard and readily available estimation tools. The paper is concluded with experimental results which support the practical value of the proposed reconstruction methodology.

Visualization and analysis of the micro-architecture of brain parenchyma by means of magnetic resonance imaging is nowadays believed to be one of the most powerful tools used for the assessment of various cerebral conditions as well as for understanding the intracerebral connectivity. Unfortunately, the conventional diffusion tensor imaging (DTI) used for estimating the local orientations of neural fibers is incapable of performing reliably in the situations when a voxel of interest accommodates multiple fiber tracts. In this case, a much more accurate analysis is possible using the high angular resolution diffusion imaging (HARDI) that represents local diffusion by its apparent coefficients measured as a discrete function of spatial orientations. In this note, a novel approach to enhancing and modeling the HARDI signals using multiresolution bases of spherical ridgelets is presented. In addition to its desirable properties of being adaptive, sparsifying, and efficiently computable, the proposed modeling leads to analytical computation of the orientation distribution functions associated with the measured diffusion, thereby providing a fast and robust analytical solution for q-ball imaging.

In this paper, we describe an approach to the problem of simultaneously enhancing image sequences and tracking the objects of interest represented by the latter. The enhancement part of the algorithm is based on Bayesian wavelet denoising, which has been chosen due to its exceptional ability to incorporate diverse a priori information into the process of image recovery. In particular, we demonstrate that, in dynamic settings, useful statistical priors can come both from some reasonable assumptions on the properties of the image to be enhanced as well as from the images that have already been observed before the current scene. Using such priors forms the main contribution of the present paper which is the proposal of the dynamic denoising as a tool for simultaneously enhancing and tracking image sequences. Within the proposed framework, the previous observations of a dynamic scene are employed to enhance its present observation. The mechanism that allows the fusion of the information within successive image frames is Bayesian estimation, while transferring the useful information between the images is governed by a Kalman filter that is used for both prediction and estimation of the dynamics of tracked objects. Therefore, in this methodology, the processes of target tracking and image enhancement "collaborate" in an interlacing manner, rather than being applied separately. The dynamic denoising is demonstrated on several examples of SAR imagery. The results demonstrated in this paper indicate a number of advantages of the proposed dynamic denoising over "static" approaches, in which the tracking images are enhanced independently of each other.

The problem of segmentation of tracking sequences is of central importance in a multitude of applications. In the current paper, a different approach to the problem is discussed. Specifically, the proposed segmentation algorithm is implemented in conjunction with estimation of the dynamic parameters of moving objects represented by the tracking sequence. While the information on objects' motion allows one to transfer some valuable segmentation priors along the tracking sequence, the segmentation allows substantially reducing the complexity of motion estimation, thereby facilitating the computation. Thus, in the proposed methodology, the processes of segmentation and motion estimation work simultaneously, in a sort of "collaborative" manner. The Bayesian estimation framework is used here to perform the segmentation, while Kalman filtering is used to estimate the motion and to convey useful segmentation information along the image sequence. The proposed method is demonstrated on a number of both computed-simulated and real-life examples, and the obtained results indicate its advantages over some alternative approaches.

In this note, we propose a method to perform segmentation on the tensor manifold, that is, the space of positive definite matrices of given dimension. In this work, we explicitly use the Riemannian structure of the tensor space in designing our algorithm. This structure has already been utilized in several approaches based on active contour models which separate the mean and/or variance inside and outside the evolving contour. We generalize these methods by proposing a new technique for performing segmentation by separating the entire probability distributions of the regions inside and outside the contour using the Bhattacharyya metric. In particular, this allows for segmenting objects with multimodal probability distributions (on the space of tensors). We demonstrate the effectiveness of our algorithm by segmenting various textured images using the structure tensor. A level set based scheme is proposed to implement the curve flow evolution equation.

Segmentation of medical ultrasound images (e.g., for the purpose of surgical or radiotherapy planning) is known to be a difficult task due to the relatively low resolution and reduced contrast of the images, as well as due to the discontinuity and uncertainty of segmentation boundaries caused by speckle noise. Under such conditions, useful segmentation results seem to be only achievable by means of relatively complex algorithms, which are usually computationally involved and/or require a prior learning. In this paper, a different approach to the problem of segmentation of medical ultrasound images is proposed. In particular, we propose to preprocess the images before they are subjected to a segmentation procedure. The proposed preprocessing modifies the images (without affecting their anatomic contents) so that the resulting images can be effectively segmented by relatively simple and computationally efficient means. The performance of the proposed method is tested in a series of both in silico and in vivo experiments.

This paper addresses the problem of image segmentation by means of active contours, whose evolution is driven by the gradient flow derived from an energy functional that is based on the Bhattacharyya distance. In particular, given the values of a photometric variable (or of a set thereof), which is to be used for classifying the image pixels, the active contours are designed to converge to the shape that results in maximal discrepancy between the empirical distributions of the photometric variable inside and outside of the contours. The above discrepancy is measured by means of the Bhattacharyya distance that proves to be an extremely useful tool for solving the problem at hand. The proposed methodology can be viewed as a generalization of the segmentation methods, in which active contours maximize the difference between a finite number of empirical moments of the "inside" and "outside" distributions. Furthermore, it is shown that the proposed methodology is very versatile and flexible in the sense that it allows one to easily accommodate a diversity of the image features based on which the segmentation should be performed. As an additional contribution, a method for automatically adjusting the smoothness properties of the empirical distributions is proposed. Such a procedure is crucial in situations when the number of data samples (supporting a certain segmentation class) varies considerably in the course of the evolution of the active contour. In this case, the smoothness properties of the empirical distributions have to be properly adjusted to avoid either over- or underestimation artifacts. Finally, a number of relevant segmentation results are demonstrated and some further research directions are discussed.

The problem of reconstruction of ultrasound images by means of blind deconvolution has long been recognized as one of the central problems in medical ultrasound imaging. In this paper, this problem is addressed via proposing a blind deconvolution method which is innovative in several ways. In particular, the method is based on parametric inverse filtering, whose parameters are optimized using two-stage processing. At the first stage, some partial information on the point spread function is recovered. Subsequently, this information is used to explicitly constrain the spectral shape of the inverse filter. From this perspective, the proposed methodology can be viewed as a "hybridization" of two standard strategies in blind deconvolution, which are based on either concurrent or successive estimation of the point spread function and the image of interest. Moreover, evidence is provided that the "hybrid" approach can outperform the standard ones in a number of important practical cases. Additionally, the present study introduces a different approach to parameterizing the inverse filter. Specifically, we propose to model the inverse transfer function as a member of a principal shift-invariant subspace. It is shown that such a parameterization results in considerably more stable reconstructions as compared to standard parameterization methods. Finally, it is shown how the inverse filters designed in this way can be used to deconvolve the images in a nonblind manner so as to further improve their quality. The usefulness and practicability of all the introduced innovations are proven in a series of both in silico and in vivo experiments. Finally, it is shown that the proposed deconvolutioh algorithms are capable of improving the resolution of ultrasound images by factors of 2.24 or 6.52 (as judged by the autocorrelation criterion) depending on the type of regularization method used.

The finite frequency bandwidth of ultrasound transducers and the nonnegligible width of transmitted acoustic beams are the most significant factors that limit the resolution of medical ultrasound imaging. Consequently, in order to recover diagnostically important image details, obscured due to the resolution limitations, an image restoration procedure should be applied. The present study addresses the problem of ultrasound image restoration by means of the blind-deconvolution techniques. Given an acquired ultrasound image, algorithms of this kind perform either concurrent or successive estimation of the point-spread function (PSF) of the imaging system and the original image. In this paper, a blind-deconvolution algorithm is proposed, in which the PSF is recovered as a preliminary stage of the restoration problem. As the accuracy of this estimation affects all the following stages of the image restoration, it is considered as the most fundamental and important problem. The contribution of the present study is twofold. First, it introduces a novel approach to the problem of estimating the PSF, which is based on a generalization of several fundamental concepts of the homomorphic deconvolution. It is shown that a useful estimate of the spectrum of the PSF can be obtained by applying a proper smoothing operator to both log-magnitude and phase of the spectra of acquired radio-frequency (RF) images. It is demonstrated that the proposed approach performs considerably better than the existing homomorphic (cepstrum-based) deconvolution methods. Second, the study shows that given a reliable estimate of the PSF, it is possible to deconvolve it out of the RF-image and obtain an estimate of the true tissue reflectivity function, which is relatively independent of the properties of the imaging system. The deconvolution was performed using the maximum a-posteriori (MAP) estimation framework for a number of statistical priors assumed for the reflectivity function. It is shown in a series of in vivo experiments that reconstructions based on the priors, which tend to emphasize the "sparseness" of the tissue structure, result in solutions of higher resolution and contrast.

In most approaches to the problem of two-dimensional homomorphic deconvolution of ultrasound images, the estimation of a corresponding point-spread function (PSF) is necessarily the first stage in the process of image restoration. This estimation is usually performed in the Fourier domain by either successive or simultaneous estimation of the amplitude and phase of the Fourier transform (FT) of the PSE This paper addresses the problem of recovering the FT-phase of the PSF, which is an important reconstruction problem by itself. The purpose of this paper is twofold. First, it provides a theoretical framework, establishing that the FT-phase of the PSF can be effectively estimated by a proper smoothing of the FT-phase of the appropriate radio-frequency (RF) image. Second, it presents a novel approach to the estimation of the FT-phase of the PSF, by solving a continuous Poisson equation over a predefined smooth subspace, in contrast to the discrete Poisson equation solver used for the classical least mean squares phase unwrapping algorithms, followed by a smoothing procedure. The proposed approach is possible due to the distinct properties of the FT-phases, among which the most important property is the availability of precise values of their partial derivatives. This property overcomes the main disadvantage of the discrete schemes, which routinely use wrapped (principal) values of the phase in order to approximate its partial derivatives. Since such an approximation is feasible subject to the restriction that the partial phase differences do not exceed pi in absolute value, the discrete schemes perform satisfactory only for few practical situations. The proposed approach is shown to be independent of this restriction and, thus, it performs for a wider class of the phases with significantly lower errors. The main advantages of the novel method over the algorithms based on discrete schemes are demonstrated in a series of computer simulations and for in vivo measurements.

A different approach to the problem of estimation of the ultrasound pulse spectrum, which usually arises as a part of ultrasound image restoration algorithms, is presented. It is shown that this estimation problem can be reformulated in terms of a de-noising problem. In this formulation, the log-spectrum of a radio-frequency line (RF-line) is viewed as a noisy measurement of the signal that needs to be estimated, i.e., the ultrasound pulse log-spectrum. The log-spectrum of the tissue reflectivity function (i.e., tissue response) is considered as the noise to be rejected. The contribution of the paper is twofold. First, it provides statistical description of the reflectivity function log-spectrum for the case, when the samples of the reflectivity function are independent identically distributed (i.i.d.) Gaussian random variables. Moreover, it is shown that the problem of the pulse spectrum recovery is essentially a de-noising problem. Consequently, it is suggested to solve the problem within the framework of the de-noising by wavelet shrinkage. Second, a computationally efficient algorithm is proposed for the pulse-spectrum estimation, which can be viewed as a modified version of the classical Donoho's three-step de-noising procedure. This modification is necessary, because of specific properties of the noise to be rejected. It is shown, that whenever the samples of the reflectivity function can be assumed to be i.i.d. Gaussian random variables, the samples of its log-spectrum obey the Fisher-Tippet distribution. For this type of noise, straightforward implementation of the standard de-noising can cause serious estimation errors. In order to overcome this difficulty, an outlier-resistant de-noising is performed. The unique properties of this modified de-noising algorithm allow estimating the pulse spectrum adaptively to its properties, as they are continuously influenced by the frequency-dependent attenuation process. The performance of the proposed algorithm is examined in a series of computer-simulations. It is shown that this algorithm, developed on the assumption of the "Gaussian" reflectivity function, remains applicable for broader classes of distributions. The results obtained in a series of in vivo experiments reveal superior performance of the novel approach over some of alternative estimation techniques, e.g., cepstrum-based estimation.

Over the last few decades there were dramatic improvements in ultrasound imaging quality with the utilization of harmonic frequencies induced by both tissue and echo-contrast agents. The advantages of harmonic imaging cause rapid penetration of this modality to diverse clinical uses, among which myocardial perfusion determination seems to be the most important application. In order to effectively employ the information, comprised in the higher harmonics of the received signals, this information should be properly extracted. A commonly used method of harmonics separation is linear filtering. One of its main shortcomings is the inverse relationship between the detectability of the contrast agent and the axial resolution. In this paper, a novel, nonlinear technique is proposed for separating the harmonic components, contained in the received radio-frequency images. It is demonstrated that the harmonic separation can be efficiently performed by means of convex optimization. It performs the separation without affecting the image resolution. The procedure is based on the concepts of sparse signal representation in overcomplete signal bases. A special type of the sparse signal representation, that is especially suitable for the problem at hand, is explicitly described. The ability of the novel technique to acquire "un-masked," second (or higher) harmonic images is demonstrated in series of computer and phantom experiments.