Magnetic Resonance Imaging (MRI) plays a vital role in diagnosis, management and monitoring of many diseases. However, it is an inherently slow imaging technique. Over the last 20 years, parallel imaging, temporal encoding and compressed sensing have enabled substantial speed-ups in the acquisition of MRI data, by accurately recovering missing lines of k-space data. However, clinical uptake of vastly accelerated acquisitions has been limited, in particular in compressed sensing, due to the time-consuming nature of the reconstructions and unnatural looking images. Following the success of machine learning in a wide range of imaging tasks, there has been a recent explosion in the use of machine learning in the field of MRI image reconstruction. A wide range of approaches have been proposed, which can be applied in k-space and/or image-space. Promising results have been demonstrated from a range of methods, enabling natural looking images and rapid computation. In this review article we summarize the current machine learning approaches used in MRI reconstruction, discuss their drawbacks, clinical applications, and current trends.
In this paper, we explore a novel method for tomographic image reconstruction in the field of SPECT imaging. Deep Learning methodologies and more specifically deep convolutional neural networks (CNN) are employed in the new reconstruction method, which is referred to as "CNN Reconstruction - CNNR". For training of the CNNR Projection data from software phantoms were used. For evaluation of the efficacy of the CNNR method, both software and hardware phantoms were used. The resulting tomographic images are compared to those produced by filtered back projection (FBP) [1], the "Maximum Likelihood …
Positron emission tomography, like many other tomographic imaging modalities, relies on an image reconstruction step to produce cross-sectional images from projection data. Detection and localization of the back-to-back annihilation photons produced by positron-electron annihilation defines the trajectories of these photons, which when combined with tomographic reconstruction algorithms, permits recovery of the distribution of positron-emitting radionuclides. Here we produce cross-sectional images directly from the detected coincident annihilation photons, without using a reconstruction algori…
In magnetic resonance imaging (MRI), inhomogeneity in the main magnetic field used for imaging, referred to as off-resonance, can lead to image artifacts ranging from mild to severe depending on the application. Off-resonance artifacts, such as signal loss, geometric distortions, and blurring, can compromise the clinical and scientific utility of MR images. In this review, we describe sources of off-resonance in MRI, how off-resonance affects images, and strategies to prevent and correct for off-resonance. Given recent advances and the great potential of low field and/or portable MRI, we also …
University of California, Berkeley（加州大学伯克利分校）；Stanford University（斯坦福大学）
Purpose: To develop a general phase regularized image reconstruction method, with applications to partial Fourier imaging, water-fat imaging and flow imaging. Theory and Methods: The problem of enforcing phase constraints in reconstruction was studied under a regularized inverse problem framework. A general phase regularized reconstruction algorithm was proposed to enable various joint reconstruction of partial Fourier imaging, water-fat imaging and flow imaging, along with parallel imaging (PI) and compressed sensing (CS). Since phase regularized reconstruction is inherently non-convex and se…
In this paper, we review the quantum mechanics of magnetic resonance imaging (MRI). We traverse its hierarchy of scales from the spin and orbital angular momentum of subatomic particles to the ensemble magnetization of tissue. And we review a number of modalities used in the assessment of acute ischemic stroke and traumatic brain injury.
Over the past few decades, magnetic resonance imaging has been utilized as a powerful imaging modality to evaluate the structure and function of various organs in the human body,such as the brain. Additionally, diffusion and perfusion MR imaging have been increasingly used in neurovascular clinical applications. In diffusion-weighted magnetic resonance imaging, the mobility of water molecules is explored in order to obtain information about the microscopic behavior of the tissues. In contrast, perfusion weighted imaging uses tracers to exploit hemodynamic status, which enables researchers and …
The properties of dephasing and the resulting relaxation of the magnetization are the basic principle on which all magnetic resonance imaging methods are based. The signal obtained from the gyrating spins is essentially determined by the properties of the considered tissue. Especially the susceptibility differences caused by magnetized materials (for example, deoxygenated blood, BOLD-effect) or magnetic nanoparticles are becoming more important for biomedical imaging. In the present work, the influence of such field inhomogeneities on the NMR-signal is analyzed.
Several successful iterative approaches have recently been proposed for parallel-imaging reconstructions of variable-density (VD) acquisitions, but they often induce substantial computational burden for non-Cartesian data. Here we propose a generalized variable-FOV PILS reconstruction 3D VD Cartesian and non-Cartesian data. The proposed method separates k-space into non-intersecting annuli based on sampling density, and sets the 3D reconstruction FOV for each annulus based on the respective sampling density. The variable-FOV method is compared against conventional gridding, PILS, and ESPIRiT r…
The uncertainty relationship in MRI is shown. The result of uncertainty relationship is compared with other factors influencing the resolution of MRI. Our estimations show that the uncertainty relationship is of no significance in practice.
Second Affiliated Hospital of Xiamen Medical College（厦门医学院附属第二医院）；Neusoft Medical System（东软医疗系统）
Magnetic resonance imaging has been widely applied in clinical diagnosis, however, is limited by its long data acquisition time. Although imaging can be accelerated by sparse sampling and parallel imaging, achieving promising reconstruction images with a fast reconstruction speed remains a challenge. Recently, deep learning approaches have attracted a lot of attention for its encouraging reconstruction results but without a proper interpretability. In this letter, to enable high-quality image reconstruction for the parallel magnetic resonance imaging, we design the network structure from the p…
In this work some possible applications of negative permeability magnetic metamaterial lenses for magnetic resonance imaging (MRI) are analyzed. Metamaterials are artificial composites designed to have a given permittivity and/or permeability, including negative values for these constants. It is shown that using magnetic metamaterials lenses it is possible to manipulate the spatial distribution of the radio-frequency (RF) field used in MR systems and, under some circumstances, improve the sensitivity of surface coils. Furthermore a collimation of the RF field, phenomenon that may find applicat…
Intel Labs（英特尔实验室）；University of California, Berkeley（加州大学伯克利分校）；Stanford University（斯坦福大学）
Magnetic resonance imaging is capable of producing volumetric images without ionizing radiation. Nonetheless, long acquisitions lead to prohibitively long exams. Compressed sensing (CS) can enable faster scanning via sub-sampling with reduced artifacts. However, CS requires significantly higher reconstruction computation, limiting current clinical applications to 2D/3D or limited-resolution dynamic imaging. Here we analyze the practical limitations to T2 Shuffling, a four-dimensional CS-based acquisition, which provides sharp 3D-isotropic-resolution and multi-contrast images in a single scan. …
In this work a mu=-1 metamaterial (MM) lens for magnetic resonance imaging (MRI) is demonstrated. MRI uses surface coils to detect the radiofrequency(RF) energy absorbed and emitted by the nuclear spins in the imaged object. The proposed MM lens manipulates the RF field detected by these surface coils, so that the coil sensitivity and spatial localization is substantially improved. Beyond this specific application, we feel that the reported results are the experimental confirmation of a new concept for the manipulation of RF field in MRI, which paves the way to many other interesting applicati…
The newly developed Core Imaging Library (CIL) is a flexible plug and play library for tomographic imaging with a specific focus on iterative reconstruction. CIL provides building blocks for tailored regularised reconstruction algorithms and explicitly supports multichannel tomographic data. In the first part of this two-part publication, we introduced the fundamentals of CIL. This paper focuses on applications of CIL for multichannel data, e.g., dynamic and spectral. We formalise different optimisation problems for colour processing, dynamic and hyperspectral tomography and demonstrate CIL's …
Purpose: To develop generic optimization strategies for image reconstruction using graphical processing units (GPUs) in magnetic resonance imaging (MRI) and to exemplarily report about our experience with a highly accelerated implementation of the non-linear inversion algorithm (NLINV) for dynamic MRI with high frame rates. Methods: The NLINV algorithm is optimized and ported to run on an a multi-GPU single-node server. The algorithm is mapped to multiple GPUs by decomposing the data domain along the channel dimension. Furthermore, the algorithm is decomposed along the temporal domain by relax…
Imaging distributions of radioactive sources plays a substantial role in nuclear medicine as well as in monitoring nuclear waste and its deposit. Coded Aperture Imaging has been proposed as an alternative to parallel or pinhole collimators, but requires image reconstruction as an extra step. Multiple reconstruction methods with varying run time and computational complexity have been proposed. Yet, no quantitative comparison between the different reconstruction methods has been carried out so far. This paper focuses on a comparison based on three sets of hot-rod phantom images captured with an …
We demonstrate an experimental method for direct 2D and 3D imaging of magnetic radio-frequency (rf) field distribution in metal-dielectric structures based on traveling wave (TW) magnetic resonance imaging (MRI) at ultra-high field (> 7T). The typical apparatus would include an ultra-high field whole body or small bore MRI scanner, waveguide elements filled with MRI active dielectrics with predefined electric and magnetic properties, and TW rf transmit-receive probes. We validated the technique by obtaining TW magnetic-resonance (MR) images of the magnetic field distribution of the rf modes of…
Purpose: To develop a rapid imaging framework for balanced steady-state free precession (bSSFP) that jointly reconstructs undersampled data (by a factor of R) across multiple coils (D) and multiple acquisitions (N). To devise a multi-acquisition coil compression technique for improved computational efficiency. Methods: The bSSFP image for a given coil and acquisition is modeled to be modulated by a coil sensitivity and a bSSFP profile. The proposed reconstruction by calibration over tensors (ReCat) recovers missing data by tensor interpolation over the coil and acquisition dimensions. Coil com…
We investigate neural network image reconstruction for magnetic particle imaging. The network performance depends strongly on the convolution effects of the spectrum input data. The larger convolution effect appearing at a relatively smaller nanoparticle size obstructs the network training. The trained single-layer network reveals the weighting matrix consisted of a basis vector in the form of Chebyshev polynomials of the second kind. The weighting matrix corresponds to an inverse system matrix, where an incoherency of basis vectors due to a low convolution effects as well as a nonlinear activ…
Magnetic Particle Imaging (MPI) is an emerging medical imaging modality that images the spatial distribution of superparamagnetic iron oxide (SPIO) nanoparticles using their nonlinear response to applied magnetic fields. In standard x-space approach to MPI, the image is reconstructed by gridding the speed-compensated nanoparticle signal to the instantaneous position of the field free point (FFP). However, due to safety limits on the drive field, the field-of-view (FOV) needs to be covered by multiple relatively small partial field-of-views (pFOVs). The image of the entire FOV is then pieced to…
Recently, metamaterials have been introduced to improve the signal-to-noise ratio (SNR) of magnetic resonance images with very promising results. However, the use polymers in the generation of high quality images in magnetic resonance imaging has not been fully been investigated. These investigations explored the use of a dielectric periodical array as a lens to improve the image SNR generated with single surface coils. Commercial polycarbonate glazing sheets were used together with a circular coil to generate phantom images at 3 Tesla on a clinical MR imager.
Medical imaging modalities have revolutionized health-care approaches by offering a better understanding of the human anatomy. Discovery of x-rays allowed the exploiting of the micro-scaled information of human anatomy. Computed tomography is one of the well-known imaging modalities using x-rays to create images of medical and non-medical objects. CT imaging has several variety of applications from medical diagnosis to industrial non-destructive testing. In this paper, we review computed tomography imaging modality and its applications in screening, diagnosis, and treatment and study some of t…
Various imaging methods for small animals are rapidly gaining acceptance in biology and medical research. Optical imaging is a very fast and convenient method of biological interrogation, but suffers from significant disadvantages, such as the absence of 3D image reconstruction algorithms. This have up until now impeded progress to a quantitative stage. Here we propose a 3D reconstruction algorithm for imaging light emitted at depths in small living animals. We believe the methods discussed here are novel and lead to a novel imaging paradigm, which we call Light Emission Tomography.
Parallel magnetic resonance imaging (MRI) is a technique of image acceleration which takes advantage of the localization of the field of view (FOV) of coils in an array. In this letter we show that metamaterial lenses based on capacitively-loaded rings can provide higher localization of the FOV. Several lens designs are systematically analyzed in order to find the structure providing higher signal-to-noise-ratio. The magnetoinductive (MI) lens is find to be the optimum structure and an experiment is developed to show it. The ability of the fabricated MI lenses to accelerate the image is quanti…
Parallel magnetic resonance imaging has served as an effective and widely adopted technique for accelerating scans. The advent of sparse sampling offers aggressive acceleration, allowing flexible sampling and better reconstruction. Nevertheless, faithfully reconstructing the image from limited data still poses a challenging task. Recent low-rank reconstruction methods exhibit superiority in providing a high-quality image. However, none of them employ the routinely acquired calibration data for improving image quality in parallel magnetic resonance imaging. In this work, an image reconstruction…
We describe an acquisition/processing procedure for image reconstruction in dynamic Magnetic Resonance Imaging (MRI). The approach requires sliding window to record a set of trajectories in the k-space, standard regularization to reconstruct an estimate of the object and compressed sensing to recover image residuals. We validated this approach in the case of specific simulated experiments and, in the case of real measurements, we showed that the procedure is reliable even in the case of data acquired by means of a low-field scanner.
Conventional Magnetic Resonance Imaging (MRI) is hampered by long scan times and only qualitative image contrasts that prohibit a direct comparison between different systems. To address these limitations, model-based reconstructions explicitly model the physical laws that govern the MRI signal generation. By formulating image reconstruction as an inverse problem, quantitative maps of the underlying physical parameters can then be extracted directly from efficiently acquired k-space signals without intermediate image reconstruction -- addressing both shortcomings of conventional MRI at the same…
Multibranch Generative Models for Multichannel Imaging with an Application to PET/CT Synergistic Reconstruction
Univ. Brest, LaTIM, INSERM, UMR 1101（布雷斯特大学、LaTIM研究所、法国国家健康与医学研究院、UMR 1101联合研究单位）；Nuclear Medicine Department, Poitiers University Hospital（普瓦捷大学医院核医学科）
Magnetic resonance (MR) imaging relies on conventional electronics that is increasingly challenged by the push for stronger magnetic fields and higher channel count. These problems can be avoided by utilizing optical technologies. As a replacement for the standard low-noise preamplifier, we have implemented a new transduction principle that upconverts an MR signal to the optical domain and imaged a phantom in a clinical 3 T scanner with signal-to-noise comparable to classical induction detection.
Molecular imaging refers to a class of noninvasive biomedical imaging techniques with the sensitivity and specificity to image biochemical variations in-vivo. An ideal molecular imaging technique visualizes a biochemical target according to a range of criteria, including high spatial and temporal resolution, high contrast relative to non-targeted tissues, depth-independent penetration into tissue, lack of harm to the organism under study, and low cost. Because no existing molecular imaging modality is ideal for all purposes, new imaging approaches are needed. Here we demonstrate a novel molecu…
The objective of this article is to introduce the tools to analyze the contrast imaging problem in Nuclear Magnetic Resonance. Optimal trajectories can be selected among extremal solutions of the Pontryagin Maximum Principle applied to this Mayer type optimal problem. Such trajectories are associated to the question of extremizing the transfer time. Hence the optimal problem is reduced to the analysis of the Hamiltonian dynamics related to singular extremals and their optimality status. This is illustrated by using the examples of cerebrospinal fluid / water and grey / white matter of cerebrum…
A central goal of modern magnetic resonance imaging (MRI) is to reduce the time required to produce high-quality images. Efforts have included hardware and software innovations such as parallel imaging, compressed sensing, and deep learning-based reconstruction. Here, we propose and demonstrate a Bayesian method to build statistical libraries of magnetic resonance (MR) images in k-space and use these libraries to identify optimal subsampling paths and reconstruction processes. Specifically, we compute a multivariate normal distribution based upon Gaussian processes using a publicly available l…
Purpose: To introduce a fast and improved direct reconstruction method for multi-shot diffusion weighted (msDW) scans for high-resolution studies. Methods:Multi-shot EPI methods can enable higher spatial resolution for diffusion MRI studies. Traditionally, such acquisitions required specialized reconstructions involving phase compensation to correct for inter-shot motion artifacts. The recently proposed MUSSELS reconstruction belongs to a new class of parallel imaging-based methods that recover artifact-free DWIs from msDW data without needing phase compensation. However, computational demands…
Breast X-ray CT imaging is being considered in screening as an extension to mammography. As a large fraction of the population will be exposed to radiation, low-dose imaging is essential. Iterative image reconstruction based on solving an optimization problem, such as Total-Variation minimization, shows potential for reconstruction from sparse-view data. For iterative methods it is important to ensure convergence to an accurate solution, since important image features, such as presence of microcalcifications indicating breast cancer, may not be visible in a non-converged reconstruction, and th…
Transmitter arrays play a critical role in ultra high field Magnetic Resonance Imaging (MRI), especially given the advantages made possible via parallel transmission (pTx) techniques. One of the challenges in design and construction of transmit arrays has traditionally been finding effective strategies for decoupling elements of the transmit array. Here, we present the design of the first self-decoupled, loop-based transmit array for human brain MRI at 10.5T / 447MHz. We demonstrate, using full-wave electromagnetic simulations, effective decoupling of the transmit elements without requiring th…
Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院保罗·C·劳特伯生物医学成像研究中心）；Research center for Medical AI, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院医学人工智能研究中心）；Nanchang University（南昌大学）；University at Buffalo, the State University of New York（纽约州立大学布法罗分校）
Medical imaging is playing a more and more important role in clinics. However, there are several issues in different imaging modalities such as slow imaging speed in MRI, radiation injury in CT and PET. Therefore, accelerating MRI, reducing radiation dose in CT and PET have been ongoing research topics since their invention. Usually, acquiring less data is a direct but important strategy to address these issues. However, less acquisition usually results in aliasing artifacts in reconstructions. Recently, deep learning (DL) has been introduced in medical image reconstruction and shown potential…
The imaging and subsequent accurate diagnosis of paediatric brain tumours presents a radiological challenge, with magnetic resonance imaging playing a key role in providing tumour specific imaging information. Diffusion weighted and perfusion imaging are commonly used to aid the non invasive diagnosis of paediatric brain tumours, but are usually evaluated by expert qualitative review. Quantitative studies are mainly single centre and single modality. The aim of this work was to combine multi centre diffusion and perfusion imaging, with machine learning, to develop machine learning based classi…
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；Massachusetts Institute of Technology（麻省理工学院）
Image corruption by motion artifacts is an ingrained problem in Magnetic Resonance Imaging (MRI). In this work, we propose a neural network-based regularization term to enhance Autofocusing, a classic optimization-based method to remove motion artifacts. The method takes the best of both worlds: the optimization-based routine iteratively executes the blind demotion and deep learning-based prior penalizes for unrealistic restorations and speeds up the convergence. We validate the method on three models of motion trajectories, using synthetic and real noisy data. The method proves resilient to n…
