Current clinical magnetic resonance (MR) acquisitions primarily rely on qualitative or 'weighted' images and diagnosis is made by subjective assessment of regional signal intensity (hyperintense or hypointense). However, MR signal for the same material can vary due to different scanners and different protocols, which hinders objective evaluation of disease severity. In contrast, quantitative MRI provides objective information for tissue characterization, offering enhanced inter-session and inter-site reproducibility. It enables improved pathology detection and disease monitoring and has better sensitivity to mild or diffuse tissue alterations compared to qualitative imaging. The combination of multiple biomarkers provides more comprehensive information and shows great promise for risk assessment and early detection.Despite these advantages, the clinical application of multiparametric MRI has been limited due to the prolonged scan times for acquiring different biomarkers, motion artifacts, and misregistration between parametric maps. MR Multitasking presents a promising approach for motion-resolved, multi-parametric mapping. However, it has yet to exploit the multi-echo information (magnitude and phase) for T2*, susceptibility, and fat fraction mapping, which necessitates further technical development. This includes flow compensation for more accurate susceptibility mapping, achieving adequate temporal resolution for motion tracking, and improving imaging efficiency for multi-echo readouts. In addition, MR Multitasking demands further improvement in quantitative performance (precision and repeatability) and scan time for practical applications. The dissertation will be focused on technical developments of MR Multitasking to enable comprehensive tissue characterization and to improve quantitative performance. The first objective is to develop a technique for three-dimensional, whole-brain simultaneous T1, T2, T2*, and susceptibility mapping. The proposed method is evaluated on phantoms and human subjects. The second objective involves further technical development to achieve free-breathing, non-ECG, simultaneous myocardial T1, T2, T2*, and FF mapping in a 2.5-min scan. Lastly, a novel reconstruction approach is introduced to improve precision and repeatability and shorten scan time. The approach is evaluated with numerical simulations and healthy subjects. The dissertation represents a step toward motion-resolved, comprehensive tissue characterization within a clinically feasible scan time and without the need for extra physiological monitoring. It lays the groundwork for future clinical use of quantitative multiparametric MRI.

In current clinical systems, magnetic resonance imaging scans for disease diagnosis and prognosis are dominated by qualitative contrast-weighted imaging. These qualitative MR images reveal regional differences in signal intensities between tissues with focal structural or functional abnormalities and tissues that are supposedly in healthy states, facilitating subjective determination for disease diagnosis. The administration of gadolinium-based contrast agents is prevalent in clinical MRI exams, which alternates the relaxation time of neighboring water protons and creates enhanced signal intensities from damaged tissues with high vascular density and thin vessel wall for better visualization. Nowadays, nearly 50% of the MRI studies were conducted with contrast agents. However, patients with renal insufficiency are at risk of developing nephrogenic system fibrosis if exposed to gadolinium-based contrast agents, and chronic toxic effects of possible gadolinium retention have been reported. In the meantime, qualitative contrast-weighted images have limited sensitivity to subtle alteration in tissue states, lack of biological specificity and multi-center reproducibility, and limited predictive values. One promising alternative is quantitative multiparametric MRI, which contains various methods to quantify multiple parameters with interpretable physical units that are intrinsic to tissue properties. Most of these quantitative approaches do not involve the administration of contrast agents, therefore ensuring the safety of the application to a wide range of patients and reducing the costs of MRI. These quantitative parameters are highly reproducible, sensitive to subtle physiological tissue changes, and specific for disease pathologies. More importantly, each of these parameters reveal tissue properties in different aspects, having the potential to offer complementary information for comprehensive tissue characterization, and acting as biomarkers that are directly associated with diseases states. Despite the benefits to clinical studies, quantitative multiparametric MRI has yet to be widely adopted in routine clinical practices because of several major technical limitations including (i) long scan times that compromises image resolution and/or spatial coverage, (ii) motion artifacts, (iii) misaligned parametric maps due to separate acquisitions, and (iv) complicated clinical workflow. This dissertation aims to address some of these challenges by proposing a simultaneous quantitative multiparametric MRI approach with Magnetic Resonance Multitasking and focus on the quantification of T1, T2, T1 , and ADC, which serves as the start of the ultimate goal to provide a clinically translatable, multiparametric whole-body quantitative tissue characterization technique. A novel approach to simultaneously quantifying T1, T2, and ADC in the brain was first developed using MR Multitasking in conjunction with a time-resolved phase correction strategy to compensate for the inter-shot phase inconsistencies introduced by physiological motion. It was implemented as a push-button, continuous acquisition that simplified the workflow. This technique was initially demonstrated in healthy subjects to efficiently produce distortion-free, co-registered T1, T2, and ADC maps with 3D brain coverage (100mm) in 9.3min. The resulting T1, T2, and ADC measurements in the brain were comparable to reference quantitative approaches. Abrupt motion was manually identified and removed to yield T1, T2, and ADC maps that were free from motion artifacts and with accurate quantitative measurements. Clinical feasibility was demonstrated on post-surgery glioblastoma patients. A motion-resolved, simultaneous T1, T2, and T1 quantification technique was then developed using MR Multitasking in a push-button 9min acquisition. Rigid intra-scan head motion was captured and simultaneously resolved along with the relaxation processes. This technique was first validated in healthy subjects to produce high quality, whole-brain (140mm) T1, T2, and T1 maps and repeatable T1, T2, and T1 measurements that were in excellent agreement with gold standard methods. Motion-resolved, artifact-free maps were generated under either in-plane or through-plane motion, which provided a novel avenue for handling rigid motion in brain MRI. Synthetic contrast-weighted qualitative images comparable to clinical images were generated using the parameter maps, demonstrating the significant potential to replace conventional MRI scans with a single Multitasking scan for clinical purposes. This technique was applied in a pilot clinical setting to perform tissue characterization in relapsing-remitting multiple sclerosis patients. The combination of T1, T2, and T1 significantly improved the accuracy of the differentiation of multiple sclerosis patients from healthy controls, compared to either single parameter alone, indicating the clinical utility of T1, T2, and T1 as quantitative biomarkers. Lastly, the above two quantitative techniques were extended to other body organs for a preliminary demonstration of potential applications, where we 1) simultaneously quantified T1, T2, and ADC in the breast with whole-breast coverage (160mm) in 8min, incorporating a B1+-compensated multiparametric fitting approach to address the notable B1+ inhomogeneity across the bilateral breast FOV, and to provide distortion-free, co-registered whole-breast T1, T2, and ADC maps with good in vivo repeatability; and 2) simultaneously quantified myocardial T1 and T1 in a single non-ECG, free-breathing acquisition, where cardiac motion and respiratory motion were retrospectively identified and simultaneously resolved to produce dynamic myocardial T1 and T1 maps of 20 cardiac phases with high temporal resolution (15ms) in a single, continuous acquisition of 1.5min per slice. Multitasking T1 and T1 measurements in the heart were comparable with gold standard techniques.

Motion Correction in MR: Correction of Position, Motion, and Dynamic Changes, Volume Eight provides a comprehensive survey of the state-of-the-art in motion detection and correction in magnetic resonance imaging and magnetic resonance spectroscopy. The book describes the problem of correctly and consistently identifying and positioning the organ of interest and tracking it throughout the scan. The basic principles of how image artefacts arise because of position changes during scanning are described, along with retrospective and prospective techniques for eliminating these artefacts, including classical approaches and methods using machine learning. Internal navigator-based approaches as well as external systems for estimating motion are also presented, along with practical applications in each organ system and each MR modality covered. This book provides a technical basis for physicists and engineers to develop motion correction methods, giving guidance to technologists and radiologists for incorporating these methods in patient examinations. Provides approaches for correcting scans prospectively and retrospectively Shows how motion and secondary effects such as field changes manifest in MR scans as artifacts and subtle biases in quantitative research Gives methods for measuring motion and associated field changes, quantifying motion and judging the accuracy of the motion and field estimates

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

This book provides an overview of current and potential applications of artificial intelligence (AI) for cardiothoracic imaging. Most AI systems used in medical imaging are data-driven and based on supervised machine learning. Clinicians and AI specialists can contribute to the development of an AI system in different ways, focusing on their respective strengths. Unfortunately, communication between these two sides is far from fluent and, from time to time, they speak completely different languages. Mutual understanding and collaboration are imperative because the medical system is based on physicians’ ability to take well-informed decisions and convey their reasoning to colleagues and patients. This book offers unique insights and informative chapters on the use of AI for cardiothoracic imaging from both the technical and clinical perspective. It is also a single comprehensive source that provides a complete overview of the entire process of the development and use of AI in clinical practice for cardiothoracic imaging. The book contains chapters focused on cardiac and thoracic applications as well more general topics on the potentials and pitfalls of AI in medical imaging. Separate chapters will discuss the valorization, regulations surrounding AI, cost-effectiveness, and future perspective for different countries and continents. This book is an ideal guide for clinicians (radiologists, cardiologists etc.) interested in working with AI, whether in a research setting developing new AI applications or in a clinical setting using AI algorithms in clinical practice. The book also provides clinical insights and overviews for AI specialists who want to develop clinically relevant AI applications.

2004 BMA Medical Book Competition Winner (Radiology category) “This is an exciting book, with a new approach to use of the MRI scanner. It bridges the gap between clinical research and general neuro-radiological practice. It is accessible to the clinical radiologist, and yet thorough in its treatment of the underlying physics and of the science of measurement. It is likely to become a classic.” British Medical Association This indispensable 'how to' manual of quantitative MR is essential for anyone who wants to use the gamut of modern quantitative methods to measure the effects of neurological disease, its progression, and its response to treatment. It contains both the methodology and clinical applications, reflecting the increasing interest in quantitative MR in studying disease and its progression. The editor is an MR scientist with an international reputation for high quality research The contributions are written jointly by MR physicists and MR clinicians, producing a practical book for both the research and medical communities A practical book for both the research and medical communities “Paul Tofts has succeeded brilliantly in capturing the essence of what needs to become the future of radiology in particular, and medicine in general – quantitative measurements of disease.” Robert I. Grossman, M.D. New York, University School of Medicine (from the Foreword)