Download or read book Model based Reconstruction of Accelerated Quantitative Magnetic Resonance Imaging MRI written by Wajiha Bano and published by . This book was released on 2019 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt:
Download or read book Quantitative Magnetic Resonance Imaging written by Nicole Seiberlich and published by Academic Press. This book was released on 2020-11-18 with total page 1094 pages. Available in PDF, EPUB and Kindle. Book excerpt: Quantitative Magnetic Resonance Imaging is a ‘go-to’ reference for methods and applications of quantitative magnetic resonance imaging, with specific sections on Relaxometry, Perfusion, and Diffusion. Each section will start with an explanation of the basic techniques for mapping the tissue property in question, including a description of the challenges that arise when using these basic approaches. For properties which can be measured in multiple ways, each of these basic methods will be described in separate chapters. Following the basics, a chapter in each section presents more advanced and recently proposed techniques for quantitative tissue property mapping, with a concluding chapter on clinical applications. The reader will learn: The basic physics behind tissue property mapping How to implement basic pulse sequences for the quantitative measurement of tissue properties The strengths and limitations to the basic and more rapid methods for mapping the magnetic relaxation properties T1, T2, and T2* The pros and cons for different approaches to mapping perfusion The methods of Diffusion-weighted imaging and how this approach can be used to generate diffusion tensor maps and more complex representations of diffusion How flow, magneto-electric tissue property, fat fraction, exchange, elastography, and temperature mapping are performed How fast imaging approaches including parallel imaging, compressed sensing, and Magnetic Resonance Fingerprinting can be used to accelerate or improve tissue property mapping schemes How tissue property mapping is used clinically in different organs Structured to cater for MRI researchers and graduate students with a wide variety of backgrounds Explains basic methods for quantitatively measuring tissue properties with MRI - including T1, T2, perfusion, diffusion, fat and iron fraction, elastography, flow, susceptibility - enabling the implementation of pulse sequences to perform measurements Shows the limitations of the techniques and explains the challenges to the clinical adoption of these traditional methods, presenting the latest research in rapid quantitative imaging which has the possibility to tackle these challenges Each section contains a chapter explaining the basics of novel ideas for quantitative mapping, such as compressed sensing and Magnetic Resonance Fingerprinting-based approaches
Download or read book Magnetic Resonance Image Reconstruction written by Mehmet Akcakaya and published by Academic Press. This book was released on 2022-11-04 with total page 518 pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic Resonance Image Reconstruction: Theory, Methods and Applications presents the fundamental concepts of MR image reconstruction, including its formulation as an inverse problem, as well as the most common models and optimization methods for reconstructing MR images. The book discusses approaches for specific applications such as non-Cartesian imaging, under sampled reconstruction, motion correction, dynamic imaging and quantitative MRI. This unique resource is suitable for physicists, engineers, technologists and clinicians with an interest in medical image reconstruction and MRI. Explains the underlying principles of MRI reconstruction, along with the latest research“/li> Gives example codes for some of the methods presented Includes updates on the latest developments, including compressed sensing, tensor-based reconstruction and machine learning based reconstruction
Download or read book MRI written by Angshul Majumdar and published by CRC Press. This book was released on 2018-09-03 with total page 222 pages. Available in PDF, EPUB and Kindle. Book excerpt: The field of magnetic resonance imaging (MRI) has developed rapidly over the past decade, benefiting greatly from the newly developed framework of compressed sensing and its ability to drastically reduce MRI scan times. MRI: Physics, Image Reconstruction, and Analysis presents the latest research in MRI technology, emphasizing compressed sensing-based image reconstruction techniques. The book begins with a succinct introduction to the principles of MRI and then: Discusses the technology and applications of T1rho MRI Details the recovery of highly sampled functional MRIs Explains sparsity-based techniques for quantitative MRIs Describes multi-coil parallel MRI reconstruction techniques Examines off-line techniques in dynamic MRI reconstruction Explores advances in brain connectivity analysis using diffusion and functional MRIs Featuring chapters authored by field experts, MRI: Physics, Image Reconstruction, and Analysis delivers an authoritative and cutting-edge treatment of MRI reconstruction techniques. The book provides engineers, physicists, and graduate students with a comprehensive look at the state of the art of MRI.
Download or read book Regularized Image Reconstruction in Parallel MRI with MATLAB written by Joseph Suresh Paul and published by CRC Press. This book was released on 2019-11-05 with total page 306 pages. Available in PDF, EPUB and Kindle. Book excerpt: Regularization becomes an integral part of the reconstruction process in accelerated parallel magnetic resonance imaging (pMRI) due to the need for utilizing the most discriminative information in the form of parsimonious models to generate high quality images with reduced noise and artifacts. Apart from providing a detailed overview and implementation details of various pMRI reconstruction methods, Regularized image reconstruction in parallel MRI with MATLAB examples interprets regularized image reconstruction in pMRI as a means to effectively control the balance between two specific types of error signals to either improve the accuracy in estimation of missing samples, or speed up the estimation process. The first type corresponds to the modeling error between acquired and their estimated values. The second type arises due to the perturbation of k-space values in autocalibration methods or sparse approximation in the compressed sensing based reconstruction model. Features: Provides details for optimizing regularization parameters in each type of reconstruction. Presents comparison of regularization approaches for each type of pMRI reconstruction. Includes discussion of case studies using clinically acquired data. MATLAB codes are provided for each reconstruction type. Contains method-wise description of adapting regularization to optimize speed and accuracy. This book serves as a reference material for researchers and students involved in development of pMRI reconstruction methods. Industry practitioners concerned with how to apply regularization in pMRI reconstruction will find this book most useful.
Download or read book Fast Quantitative Magnetic Resonance Imaging written by Guido Buonincontri and published by Springer Nature. This book was released on 2022-05-31 with total page 124 pages. Available in PDF, EPUB and Kindle. Book excerpt: Among medical imaging modalities, magnetic resonance imaging (MRI) stands out for its excellent soft-tissue contrast, anatomical detail, and high sensitivity for disease detection. However, as proven by the continuous and vast effort to develop new MRI techniques, limitations and open challenges remain. The primary source of contrast in MRI images are the various relaxation parameters associated with the nuclear magnetic resonance (NMR) phenomena upon which MRI is based. Although it is possible to quantify these relaxation parameters (qMRI) they are rarely used in the clinic, and radiological interpretation of images is primarily based upon images that are relaxation time weighted. The clinical adoption of qMRI is mainly limited by the long acquisition times required to quantify each relaxation parameter as well as questions around their accuracy and reliability. More specifically, the main limitations of qMRI methods have been the difficulty in dealing with the high inter-parameter correlations and a high sensitivity to MRI system imperfections. Recently, new methods for rapid qMRI have been proposed. The multi-parametric models at the heart of these techniques have the main advantage of accounting for the correlations between the parameters of interest as well as system imperfections. This holistic view on the MR signal makes it possible to regress many individual parameters at once, potentially with a higher accuracy. Novel, accurate techniques promise a fast estimation of relevant MRI quantities, including but not limited to longitudinal (T1) and transverse (T2) relaxation times. Among these emerging methods, MR Fingerprinting (MRF), synthetic MR (syMRI or MAGIC), and T1‒T2 Shuffling are making their way into the clinical world at a very fast pace. However, the main underlying assumptions and algorithms used are sometimes different from those found in the conventional MRI literature, and can be elusive at times. In this book, we take the opportunity to study and describe the main assumptions, theoretical background, and methods that are the basis of these emerging techniques. Quantitative transient state imaging provides an incredible, transformative opportunity for MRI. There is huge potential to further extend the physics, in conjunction with the underlying physiology, toward a better theoretical description of the underlying models, their application, and evaluation to improve the assessment of disease and treatment efficacy.
Download or read book Development and Optimization of Methods for Accelerated Magnetic Resonance Imaging written by Tom Hilbert and published by . This book was released on 2018 with total page 111 pages. Available in PDF, EPUB and Kindle. Book excerpt: Mots-clés de l'auteur: magnetic resonance imaging; quantitative imaging; acquisition acceleration; model-based reconstruction.
Download or read book Model based Reconstruction of Magnetic Resonance Spectroscopic Imaging written by Itthi Chatnuntawech and published by . This book was released on 2013 with total page 80 pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic resonance imaging (MRI) is a medical imaging technique that is used to obtain images of soft tissue throughout the body. Since its development in the 1970s, MRI has gained tremendous importance in clinical practice because it can produce high quality images of diagnostic value in an ever expanding range of applications from neuroimaging to body imaging to cancer. By far the dominant signal source in MRI is hydrogen nuclei in water. The presence of water at high concentration (-50M) in body tissue, combined with signal contrast modulation induced by the local environment of water molecules, accounts for the success of MRI as a medical imaging modality. As opposed to conventional MRI, which derives its signal from the water component, magnetic resonance spectroscopy (MRS) acquires the magnetic resonance signal from other chemical components, most frequently various metabolites in the brain, but also signals from tumors in breast and prostate. The spectroscopic signal arises from low concentration (-1 - 10mM) compounds, but in spite of the challenges posed by the resulting low signal-to-noise ratio (SNR), the development of MRS is motivated by the desire to directly observe signal sources other than water. The combination of MRS with spatial encoding is called magnetic resonance spectroscopic imaging (MRSI). MRSI captures not only the relative intensities of metabolite signals at each voxel, but also their spatial distributions. While MRSI has been proven to be clinically useful, it suffers from fundamental tradeoffs due to the inherently low SNR, such as long acquisition time and low spatial resolution. In this thesis, techniques that combine benefits from both model-based reconstruction methods and regularized reconstructions with prior knowledge are proposed and demonstrated for MRSI. These methods address constraints on acquisition time in MRSI by undersampling data during acquisition in combination with improved image reconstruction methods.
Download or read book Compressed Sensing for Magnetic Resonance Image Reconstruction written by Angshul Majumdar and published by Cambridge University Press. This book was released on 2015-02-26 with total page 228 pages. Available in PDF, EPUB and Kindle. Book excerpt: Expecting the reader to have some basic training in liner algebra and optimization, the book begins with a general discussion on CS techniques and algorithms. It moves on to discussing single channel static MRI, the most common modality in clinical studies. It then takes up multi-channel MRI and the interesting challenges consequently thrown up in signal reconstruction. Off-line and on-line techniques in dynamic MRI reconstruction are visited. Towards the end the book broadens the subject by discussing how CS is being applied to other areas of biomedical signal processing like X-ray, CT and EEG acquisition. The emphasis throughout is on qualitative understanding of the subject rather than on quantitative aspects of mathematical forms. The book is intended for MRI engineers interested in the brass tacks of image formation; medical physicists interested in advanced techniques in image reconstruction; and mathematicians or signal processing engineers.
Download or read book Accelerating Pediatric Magnetic Resonance Imaging Using Deep Learning based Image Reconstruction written by Christopher Michael Sandino and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic resonance imaging (MRI) is a powerful diagnostic tool for visualizing soft tissue anatomy, but physical limits on data acquisition speed result in uncomfortably long MRI exams. This is problematic for many patient populations, but especially for pediatric patients, who often require general anesthesia (GA) to reduce anxiety and body motion. Many attempts at accelerating data acquisition have been made to reduce or eliminate use of GA for pediatric MRI. For example, compressed sensing (CS) methods have been used to iteratively reconstruct rapidly acquired measurements into high-quality images by leveraging sparse priors. More recently, deep learning (DL) methods have been used to train deep neural network models to map the rapidly acquired measurements into even higher-quality images. While DL reconstruction approaches may potentially accelerate data acquisition beyond CS, these approaches have several issues which impede their clinical adoption. First, DL reconstructions require large quantities of high-quality ground truth data for supervised training, which can be costly and time-consuming to acquire. Second, memory requirements during network training limit the applicability of DL reconstruction to low-dimensional MRI data, such as static or dynamic 2-D imaging with limited spatiotemporal resolution. In this thesis, a series of projects demonstrating robust DL reconstruction techniques for acceleration of high-dimensional pediatric MRI will be presented. First, physics-based models are incorporated into deep CNN architectures to enforce consistency between intermediate network outputs and the rapidly acquired measurements in a novel method called DL-ESPIRiT. Physics-based modeling allows DL-ESPIRiT to be trained end-to-end in a supervised fashion with relatively little training data compared to non-physics-driven DL reconstruction. DL-ESPIRiT is applied and validated on 12X prospectively accelerated dynamic 2-D MRI scans acquired at Lucile Packard Children's Hospital. Finally, DL-ESPIRiT is extended to leverage subspace methods within the network to address GPU memory limitations during training. This method, known as deep learning-based subspace reconstruction (DL-Subspace), reconstructs a compressed representation of the MRI data instead of the data directly, thereby reducing the memory footprint during training and accelerating DL inference times. DL-Subspace is demonstrated to reconstruct 2-D dynamic MRI data with 4X higher memory efficiency and inference speed.
Download or read book Reconstruction Methods for Accelerated Magnetic Resonance Imaging written by Tao Zhang and published by . This book was released on 2014 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic resonance imaging (MRI) is a powerful medical imaging modality widely used in clinical practice. MRI provides excellent soft-tissue contrast, and does not involve ionizing radiation. In an ideal clinical setting for MRI, several requirements have to be met. First of all, diagnostic image quality has to be achieved. Second, fast image reconstruction is required, so that the radiologists can review the images before releasing the patients. Third, fast data acquisition is desired. Short scan time can not only improve patient comfort, but also reduce many imaging artifacts and improve image quality. While advanced methods, such as parallel imaging and compressed sensing, can accelerate MRI data acquisition to some extent, the achievable scan time is still very limited for several MR applications. Meanwhile, the reconstruction time for these advanced methods can take up to hours, and become clinically infeasible. This dissertation describes approaches to maintain a clinically feasible reconstruction time for advanced reconstructions, and approaches to further accelerate MRI applications, specifically MR parameter mapping and dynamic contrast-enhanced (DCE) MRI. The ultimate goal of this work is to make MRI more clinically practical. To maintain a clinically feasible reconstruction time for advanced reconstructions with large coil arrays, a geometric-decomposition coil compression method is proposed. The proposed method exploits the spatially varying data redundancy of large coil arrays, and can compress the raw data from original coils into very few virtual coils. The advanced reconstruction can be directly performed on the virtual coils instead of the original coils. The reconstruction time for large 3D datasets, acquired with 32-channel coils and reconstructed by a combined parallel imaging compressed sensing method, can be reduced to under a minute. The proposed method has been implemented in Lucile Packard Children's Hospital at Stanford. The clinical evaluation suggests that the proposed method can achieve very fast reconstruction without compromising overall image quality and delineation of anatomical structures. MR parameter mapping is a promising approach to characterize intrinsic tissue-dependent information. To accelerate lengthy MR parameter mapping, which can take up to half an hour or more, a locally low-rank method has been proposed. The proposed method has been combined with parallel imaging to achieve further acceleration. Based on preliminary result, the combined parallel imaging locally low-rank method can accelerate variable flip angle T1 mapping by factor of 6, without obvious imaging artifacts. DCE MRI is a standard component of abdominal MRI exams, most commonly used to detect and characterize mass lesions and assess renal function. 3D DCE MRI is often limited compromised spatiotemporal resolution and motion artifacts. In this work, a combined locally low-rank parallel imaging method with soft gating is proposed. The proposed method can significantly reduce motion artifacts for completely free-breathing acquisition and remove the need for deep anesthesia. The high spatiotemporal resolution achieved by the proposed method can also capture the rapid contrast hemodynamics. The proposed method has been deployed clinically in Lucile Packard Children's Hospital at Stanford. Preliminary clinical evaluation results suggest that the proposed method can achieve an image quality very close to a respiratory-triggered data acquisition, but with much higher spatiotemporal resolution.
Download or read book Principles of Magnetic Resonance Imaging written by Zhi-Pei Liang and published by Wiley-IEEE Press. This book was released on 2000 with total page 442 pages. Available in PDF, EPUB and Kindle. Book excerpt: In 1971 Dr. Paul C. Lauterbur pioneered spatial information encoding principles that made image formation possible by using magnetic resonance signals. Now Lauterbur, "father of the MRI", and Dr. Zhi-Pei Liang have co-authored the first engineering textbook on magnetic resonance imaging. This long-awaited, definitive text will help undergraduate and graduate students of biomedical engineering, biomedical imaging scientists, radiologists, and electrical engineers gain an in-depth understanding of MRI principles. The authors use a signal processing approach to describe the fundamentals of magnetic resonance imaging. You will find a clear and rigorous discussion of these carefully selected essential topics: Mathematical fundamentals Signal generation and detection principles Signal characteristics Signal localization principles Image reconstruction techniques Image contrast mechanisms Image resolution, noise, and artifacts Fast-scan imaging Constrained reconstruction Complete with a comprehensive set of examples and homework problems, Principles of Magnetic Resonance Imaging is the must-read book to improve your knowledge of this revolutionary technique.
Download or read book Deep Learning Based Image Reconstruction in Abdominal and Cardiac Magnetic Resonance Imaging MRI written by Chang Gao and published by . This book was released on 2023 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: MRI plays an important role in abdominal and cardiac imaging due to its excellent soft tissue contrast and high image resolution. Despite the benefit of excellent image quality, MRI acquisition is intrinsically slow, causing patient discomfort and slowing down the clinical workflow, which hinders its broad clinical use. For decades, undersampling reconstruction techniques have been investigated to accelerate MRI acquisition. Traditional parallel imaging and compressed sensing methods either have limited acceleration capability or require extensive computational and time resources. While the recent development of deep learning achieved unprecedented performance in image reconstruction and image enhancement tasks, there are challenges remaining to be solved. One challenge is the potential loss of image details due to network over-smooths or over-regularization. Another challenge is that networks may struggle to generalize well to diverse MRI data acquired under different conditions. In medical imaging, high-quality diverse datasets are challenging to acquire, especially for rare or specialized MRI applications. Lastly, for non-Cartesian sampling, the reconstruction can be challenging due to the need for time-consuming interpolation of non-Cartesian k-space onto a Cartesian basis. The overall goal of the dissertation is to contribute to the development of deep learning-based accelerated image reconstruction techniques and investigate the challenges in network development as mentioned above. Specifically, we aim to develop deep learning networks to improve image quality and reduce artifacts and noise for the application of (1) undersampled radial MRI reconstruction in the abdomen (aims 1 and 2), and (2) ferumoxytol-enhanced cardiac cine MRI reconstruction (aim 3). In aim 1, I developed a generative adversarial network using paired undersampled and ground truth images to reduce streaking artifacts and preserve image sharpness. In aim 2, I developed a radial k-space prediction framework by training an attention-based transformer network on k-space data. By combining the acquired and predicted k-space data, the reconstructed images will have an improved signal-to-noise ratio and fewer streaking artifacts. In aim 3, I developed an unrolled spatiotemporal deep learning network for ferumoxytol-enhanced cardiac cine MRI reconstruction. The network was trained using non-contrast-enhanced bSSFP cine images and can be successfully generalized to ferumoxytol-enhanced images.
Download or read book Model based Reconstruction for Correcting Magnetic Susceptibility induced Artifacts in Magnetic Resonance Imaging written by Giang-Chau Ngo and published by . This book was released on 2018 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt:
Download or read book Development of Deep Learning Methods for Magnetic Resonance Imaging Reconstruction and Analysis written by Yuhua Chen and published by . This book was released on 2021 with total page 187 pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic resonance image (MRI) is a widely used non-invasive radiation-free imaging technique that uniquely provides structural and functional information for disease detection, diagnosis, and treatment planning. However, the conventional MRI imaging techniques are typically slow and low in spatial or time resolution, resulting in long scan times and more susceptibility to motion artifacts. Moreover, a fast MRI scan usually comes in a low spatial resolution, making it less desirable for clinical application. A recently proposed technique, Multi-tasking MRI (MTMRI), significantly improves the scan efficiency with high temporal resolution. Nevertheless, the iterative reconstruction requires a lot of computational resources and takes a long time to process, making it challenging to fit in the clinical routine. Additionally, when doing image post-processing with MRI, despite MRI providing a good contrast of soft tissues, the variety in weighted contrast MRI's intensity values makes it challenging to extract image features compared with other quantitative imaging techniques. The most significant contribution of this dissertation's work is to address the three limitations above by developing a unified multi-purpose structure with deep-learning (DL) techniques. We achieved three primary goals in three different areas: 1) A general framework for highly accelerated MRI scanning without sacrificing spatial resolution, 2) reduce reconstruction time for motion-resolved free-breathing MRI technique, 3) accurately fully automated segmentation for abdominal MRI for fast image post-processing. All technical improvements utilize DL techniques to improve MRI in different aspects: to improve image quality in fast MRI scans, reduce reconstruction time in motion-resolved MRI, and reduce tedious human labors in abdominal MRI. First, a DL-based Super-Resolution (SR) technique is developed and evaluated in both brain MRI and coronary MR Angiography (MRA). SR can recover the image quality and structural details from a 4x and 16x low-resolution fast MRI scan. For brain MRI, several SR networks have been developed. The proposed network (mDCSRN) has successfully recovered the brain structural details from a 4x low-resolution fast scan. It is developed and evaluated on an open access high-resolution T1w brain MRI with 1131 healthy volunteers. Quantitative results show that it can achieve 4x acceleration in scan while keeping similar image quality. For coronary MRA, introducing a domain adaptive network (DRAGAN) jointly trained on both coronary and brain MRA to overcome catastrophic failures commonly in training a GAN in a small dataset, we successfully accelerated the MRA acquisition by a factor of 16. Second, DL networks are developed to accelerate the reconstruction of a 5-dimensional (5D) Multitasking MRI (MTMRI). The MTMRI is a respiratory and cardiac-motion-resolved, high-temporal-resolution technique that provides quantitative T1 mapping. However, the massive size of many dynamic MRI problems prevents deep learning networks from directly exploiting global temporal relationships. By applying deep neural networks inside a priori calculated temporal feature spaces, we enable deep learning reconstruction with global temporal modeling even for image sequences with >40,000 frames. One proposed variation of our approach using dilated multi-level Densely Connected Network (mDCN) speeds up feature space coordinate calculation by 3000x compared to conventional iterative methods, from 20 minutes to 0.39 seconds. Thus, the combination of low-rank tensor and deep learning models makes large-scale dynamic MRI feasible and practical for routine clinical application. Third, we developed Automated deep Learning-based Abdominal Multi-Organ segmentation (ALAMO) technique based on 2D U-net and a densely connected network structure with tailored design in data augmentation and training procedures. The model takes in multi-slice MR images and generates the output of segmentation results. 3.0-Tesla T1 VIBE (Volumetric Interpolated Breath-hold Examination) images of 102 subjects were used in our study. Ten OARs were studied, including the liver, spleen, pancreas, left/right kidneys, stomach, duodenum, small intestine, spinal cord, and vertebral bodies. ALAMO generated segmentation labels in good agreement with the manual results. Specifically, among the 10 OARs, 9 achieved high Dice Similarity Coefficients (DSCs) in the range of 0.87-0.96, except for the duodenum with a DSC of 0.80. Overall, the ALAMO model matched the state-of-the-art techniques in performance.
Download or read book Automating and Accelerating Magnetic Resonance Imaging written by Ke Lei and published by . This book was released on 2023 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Magnetic resonance imaging (MRI) can provide high-quality multi-contrast diagnostic images. It is non-invasive and does not use ionizing radiation. Therefore, it is safe for young patients. MRI exams follow a procedure consisting of preparation, scan prescription, data collection (i.e., the actual scanning), image reconstruction, and checking the result. Unfortunately, MRI has significantly longer scan times compared to other modalities such as computed tomography (CT). These long scan times are especially challenging for children who may struggle to stay still. In this dissertation, we aim to expedite the whole MRI exam procedure by automating and accelerating four parts of the scanning process: prescription, data collection, reconstruction, and the after-scan check. This is done through a series of three projects. First, we present a method for region-of-interest (ROI) prediction and field-of-view (FOV) prescription. Manual prescription of the field of view by MRI technologists is variable and prolongs the scanning process. Often, the FOV is either too large or crops critical anatomy. We propose a deep-learning framework, trained with radiologists' supervision, for automating FOV prescription. An intra-stack shared feature extraction network and an attention network are used to process a stack of 2D image inputs to generate scalars defining the location of a rectangular ROI. The attention mechanism is used to make the model focus on a small number of informative slices in a stack. Then the smallest FOV that makes the neural network predicted an ROI free of aliasing is calculated by an algebraic operation derived from MR sampling theory. The framework's performance is examined quantitatively with intersection over union (IoU) and pixel error on position, and qualitatively with a reader study. The framework's prescription is clinically acceptable 92\% of the time as rated by an experienced radiologist. Second, we present a learning-based model for reconstructing undersampled data using unpaired adversarial training. The lack of ground-truth MR images impedes the common supervised training of neural networks for image reconstruction. To cope with this challenge, this work leverages unpaired adversarial training for reconstruction networks, where the inputs are undersampled k-space data and naively reconstructed images from one dataset, and the labels are high-quality images from another dataset. The reconstruction networks consist of a generator which suppresses the input image artifacts, and a discriminator using a pool of (unpaired) labels to adjust the reconstruction quality. The generator is an unrolled neural network -- a cascade of convolutional and data consistency layers. The discriminator is also a multilayer Convolutional Neural Network (CNN) that plays the role of a critic scoring the quality of reconstructed images based on the Wasserstein distance metric. Our experiments with knee MRI datasets demonstrate that the proposed unpaired training enables diagnostic-quality reconstruction when high-quality image labels are not available, or when the amount of label data is small. In addition, our adversarial training scheme can achieve better image quality (as rated by expert radiologists) compared with the paired training methods using pixel-wise loss. Finally, we present a no-reference image quality assessment (IQA) framework that checks the exam outcome. In clinical practice MR images are often first seen by radiologists long after the scan. If image quality is inadequate the patient may have to return for an additional scan, or a suboptimal interpretation is rendered. Automatic IQA would enable real-time remediation. Existing IQA methods for MRI give only a general quality score. These are agnostic to the cause of the low-quality scan and the solution for improvement. Furthermore, radiologists' image quality requirements vary with the scan type and diagnostic task. Therefore, the same score may have different implications for different scans. We propose a framework with a multi-task CNN model trained with calibrated labels and measured with image rulers. Labels calibrated by human inputs follow a well-defined and efficient labeling task. Image rulers address varying quality standards and provide a concrete way of interpreting raw scores from the CNN. The model supports assessments of perceptual noise level, rigid motion, and peristaltic motion. Our experiments show that label calibration, image rulers, and multi-task training improve the model's performance and ability to generalize.
Download or read book Compressed Sensing Magnetic Resonance Image Reconstruction Algorithms written by Bhabesh Deka and published by Springer. This book was released on 2018-12-29 with total page 122 pages. Available in PDF, EPUB and Kindle. Book excerpt: This book presents a comprehensive review of the recent developments in fast L1-norm regularization-based compressed sensing (CS) magnetic resonance image reconstruction algorithms. Compressed sensing magnetic resonance imaging (CS-MRI) is able to reduce the scan time of MRI considerably as it is possible to reconstruct MR images from only a few measurements in the k-space; far below the requirements of the Nyquist sampling rate. L1-norm-based regularization problems can be solved efficiently using the state-of-the-art convex optimization techniques, which in general outperform the greedy techniques in terms of quality of reconstructions. Recently, fast convex optimization based reconstruction algorithms have been developed which are also able to achieve the benchmarks for the use of CS-MRI in clinical practice. This book enables graduate students, researchers, and medical practitioners working in the field of medical image processing, particularly in MRI to understand the need for the CS in MRI, and thereby how it could revolutionize the soft tissue imaging to benefit healthcare technology without making major changes in the existing scanner hardware. It would be particularly useful for researchers who have just entered into the exciting field of CS-MRI and would like to quickly go through the developments to date without diving into the detailed mathematical analysis. Finally, it also discusses recent trends and future research directions for implementation of CS-MRI in clinical practice, particularly in Bio- and Neuro-informatics applications.