published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

No description available.

No description available.

Diss. Linköping : Univ., 2003.

Introduction There is a pressing clinical need to treat patients with acute coronary syndrome (ACS) timely and efficiently in order to improve their prognosis. Standard tools available in ED, while useful, do not comprehensively characterize ACS for either diagnosis or risk stratification. The role of CMR in ACS is emerging because it allows assessment of both myocardial composition and function. Newer CMR techniques such as: a) T2 W imaging for assessing myocardial oedema and area at risk B) pre contrast T1 mapping techniques for quantitative characterization of the tissue composition, are adding further utility for CMR in ACS. At present the clinical use of these techniques is still limited and further investigations are needed to assess their clinical applicability in ACS patients. Aims The aims of this thesis were several. Firstly we sought to establish a CMR protocol for imaging ACS patients on a 3T CMR scanner. In order to do so, we validated a novel T2 W technique for oedema imaging (T2 prep SSFP) at 3T. Second, we aimed to perform a detailed study of the time course of oedema in ACS patients in order to establish the appropriate imaging time for the assessment of area at risk. Third, by applying T2W acute oedema imaging, we sought to investigate the functional and pathological meaning of complicated remote plaques in patients with multivessel disease. Finally, we aimed to establish whether, in comparison to standard CMR techniques, novel precontrast Tl mapping allows better characterisation of the acutely injured myocardium and whether it can predict long-term functional recovery. Methods The research studies were all performed on a 3T Trio Siemens scanner. In the initial stage of the research, we validated the T2 W technique performing phantom work and scanning both volunteers and patients to assess the uniformity of signal intensity in the myocardium and to establish a threshold based method to post process the images. We then established a CMR protocol for ACS including oedema imaging, T1 mapping imaging, perfusion, functional and late gadolinium enhancement imaging. Patients with acute myocardial infarction (both ST elevation myocardial infarction (STEMI) and non STEMI) were scanned at 4 different time points after the acute event (3 scans within 2 weeks and one at 6 months). All STEMI patients underwent primary percutaneous coronary intervention (PCI) while the non-STEMI patients underwent coronary angiography and for PCI. Results We validated the T2prep SSFP technique at 3T, highlighting its limitations and establishing a threshold of mean ± 2SD to assess myocardial oedema. We found that the optimal imaging window to assess the maximal expression of myocardial oedema was within 1 week from the acute event in patients with ST elevation MI. Also, our results showed a reduction of LGE over time (from acute to chronic) in segments which also showed improvement in contractile function indicating that even segments with transmural LGE assessed in the early hours post event could be viable. By applying these techniques in acute patients with bystander disease undergoing percutaneous coronary intervention, we found that: l) T2W imaging can detect myocardial injury downstream from a vessel identified as "non culprit" 2) in 20% of NSTEMI patients, the angiographic assessment alone failed to identify the culprit vessel. Finally, we found that the diagnostic performance of acute pre-contrast Tl-mapping was at least as good as that ofT2W CMR for detecting myocardial injury. There was a significant relationship between the segmental damaged fraction assessed by either by LGE or T2W, and mean segmental Tl values and the likelihood of improvement of segmental function at 6 months decreased progressively as acute Tl values increased. Conclusions In summary, we defined a stable imaging window for the retrospective evaluation of area at risk and we also indicated that acutely detected LGE does not necessarily equate with irreversible injury and may severely underestimate salvaged myocardium. Furthermore, in NSTEMI patients with multivessel disease, by revealing acute myocardial damage in territories pertaining to vessels not treated acutely, we raised the issue of the need for better tools for the correct identification of the culprit vessel and to stratify patients rather than by angiographic assessment alone. Finally, we demonstrated how pre-contrast Tl mapping allows for assessment of the extent of myocardial damage and how Tl mapping might become an important complementary technique to LGE and T2W for the identification of reversible myocardial injury and the prediction of functional recovery in acute MI.

No description available.

Luo Guo. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2003. / Includes bibliographical references (leaves 97-109). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.2 / Chapter 1.2 --- Basics --- p.3 / Chapter 1.2.1 --- Anatomy of Human Heart --- p.3 / Chapter 1.2.2 --- The Philosophy of MRI --- p.5 / Chapter 1.2.3 --- MRI in Practice --- p.7 / Chapter 1.3 --- Cardiac MR Images Analysis --- p.7 / Chapter 1.3.1 --- Heart Boundary Segmentation --- p.7 / Chapter 1.3.2 --- Motion Reconstruction --- p.13 / Chapter 1.4 --- Summary and Thesis Overview --- p.17 / Chapter 2 --- Tracking Tags in SPAMM Images --- p.21 / Chapter 2.1 --- Introduction --- p.21 / Chapter 2.2 --- The Snake Model --- p.28 / Chapter 2.3 --- The Improved Snake Model: Tracking Tags Using Snakes --- p.30 / Chapter 2.3.1 --- Imaging Protocol --- p.30 / Chapter 2.3.2 --- Model Formulation --- p.31 / Chapter 2.3.3 --- Numerical Solution --- p.39 / Chapter 2.4 --- Experimental Results --- p.44 / Chapter 3 --- B-Spline Based LV Motion Reconstruction --- p.52 / Chapter 3.1 --- Introduction --- p.52 / Chapter 3.2 --- LV Shape: Generalized Deformable Ellipsoid --- p.56 / Chapter 3.3 --- The New Geometric Model: Generalized Prolate Spheroid --- p.58 / Chapter 3.3.1 --- Generalized Prolate Spheroid --- p.58 / Chapter 3.3.2 --- Initial Geometric Fitting --- p.59 / Chapter 3.4 --- Fast Motion Reconstruction: The Enhanced Hi- erarchical Motion Decomposition --- p.65 / Chapter 3.4.1 --- Hierarchical Motion Decomposition --- p.65 / Chapter 3.4.2 --- Motion Reconstruction --- p.68 / Chapter 3.4.3 --- Implementation --- p.76 / Chapter 3.4.4 --- Time Smoothing --- p.77 / Chapter 3.5 --- Experimental Results --- p.79 / Chapter 3.5.1 --- Geometric Fitting --- p.79 / Chapter 3.5.2 --- Motion Reconstruction --- p.79 / Chapter 4 --- Conclusion --- p.93 / Bibliography --- p.109

Thesis (Ph. D.)--Ohio State University, 2008. / Title from first page of PDF file. Includes bibliographical references (p. 102-122).

High image fidelity in Magnetic Resonance Imaging (MRI) relies on precise magnetic field control of encoding gradient fields and background B0 magnetic fields. To ensure a high degree of accuracy in the spatial location of the proton spins and the resultant object geometry, conventional image encoding using linear gradient fields, as well as advanced techniques with non-linear encoding, requires field generating hardware capable of excellent field shaping capabilities and accuracy. Non-homogeneous B0 background fields in MR imaging cause faster relaxation, signal dropout, and geometry distortion, resulting in inferior image quality and reduced diagnostic accuracy. Besides manufacturing imperfections in the magnet and site conditions, the magnetic field inside the imaging object is not homogeneous due to the differences in geometries and magnetic properties of individual human tissues, which is recognized as the primary source of B0 variation in MRI. Considering the differences of B0 conditions across subjects, it is essential for MR imaging to utilize flexible B0 shimming techniques such as active shimming in order to produce a highly homogeneous B0 field. The control capability and optimized control strategy for these magnetic fields require the development of new hardware and methodologies. B0 background field generated by the magnet and the encoding gradient field from gradient coil are two critical pillars of MR imaging. Since the multi-coil array provides advanced shim capability and is proven to be capable of imaging encoding with a compact size, it is considered a perfect component as a combination of B0 shim coil and encoding gradient coil for an accessible head-only MR scanner. MR scanners like this type provide unique features that will enable researchers to develop new MRI methodologies and conduct research into the functionalities of the human brain through more natural human behaviors. Its clinical applications will be more accessible to the general population for disease screening and diagnosis due to its portability and low energy requirements. Since the multi-coil array has the advantage of smaller volume and wall thickness than the traditional gradient coil, its design and implementation is challenging because of its compact space, irregular curved shape of coil elements, mechanical reliability requirements during scan and good thermal control for long working periods. It was the challenges involved in the design and implementation of the multi-coil array that initiated the first project of my dissertation. In this project, we present 1) a novel molding method for the construction of resin-impregnated wire patterns with irregular curved shapes along with a microcontroller-driven motorized machine for automated coil construction, 2) the design and validation of a water-cooling system using multiple parallel pipes impregnated with thermal epoxy, 3) a quality-controlled procedure of building the multi-coil array employing the technique of vacuum resin infusion. A multi-coil array was fabricated successfully and evaluated in multiple sites and then integrated into the first-prototype of the accessible head-only MR scanner. The similar quality of experimental images from the fabricated multi-coil array compared to those from conventional gradient coils indicates that the multi-coil array can effectively shape fields for both image encoding and B0 shimming. Our lab has shown that multi-coil technology offers advanced shim capability when imaging the human brain, but it could potentially benefit the imaging of other organs like the heart. The MR imaging of the heart is subject to dark band artifacts or signal loss caused by B0 inhomogeneity, which can result in misinterpretation of lesions and a reduction in diagnostic accuracy. It has been demonstrated in a recent study that the use of multi-coil techniques can significantly reduce B0 inhomogeneity within the heart based on shim analysis using in vivo B0 maps. Multi-coil arrays are not a standard configuration in commercial scanners but are normally used for research, B0 shimming is typically implemented by using the commonly-installed spherical harmonic shim coils in the first, second, and potentially third orders. The development of multi-coil technology, more in-depth design of the coil structure and geometry as well as the optimal use of the current spherical harmonic shim technology require a thorough understanding of cardiac B0 conditions across subjects and at a population level. Since the in vivo cardiac B0 measurement is not a routine clinical protocol and dedicated in vivo measurement for a large sample size are extremely labor intensive and expensive, the lack of such B0 data is a long-standing problem, especially for the subject groups like pediatric or elderly patients who cannot undergo B0 map measurement with breath hold. This challenge could be resolved by the use of B0 simulation on the basis of structural images from different imaging modalities, assuming that the B0 distributions inside the human heart depends on the anatomical structures surrounding heart and across the entire body. The challenge and assumption led to my second project regarding B0 magnetic field simulation in the human heart. We proposed a novel B0 simulation approach based on chest-abdomen-pelvis structural CT images and validated it using in vivo acquired B0 maps in the heart from the same subjects. This B0 simulation approach was then applied to CT images from more than one thousand subjects and the resultant large set of simulated B0 maps were analyzed with different shim types for searching optimal shim solution based on popular spherical harmonic decomposition. The derived B0 conditions were also statistically analyzed for potential correlation and linear association with demographic parameters of these subjects for investigating potential population-based shim strategy. By the use of in vivo acquisition, we also investigated the B0 magnetic field variation across cardiac cycle and evaluated the impact of these variations on in vivo cardiac B0 shimming. The results of this study allow us to better understand the primary sources and characteristics of B0 distributions in the heart as well as pave the way for developing optimal B0 shim methods within heart in both subject-specific and population-based manners.

Cardiac Resynchronization Therapy (CRT) is a treatment option for heart failure patients with ventricular dyssynchrony. CRT corrects for dyssynchrony by electrically stimulating the septal and lateral walls of the left ventricle (LV), forcing synchronous con- traction and improving cardiac output. Current selection criteria for CRT rely upon the QRS duration, measured from a surface electrocardiogram, as a marker of electrical dyssynchrony. Unfortunately, 30-40% of patients undergoing CRT fail to benefit from the treatment. A multitude of studies have shown that presence of mechanical dyssynchrony in the LV is an important factor in determining if a patient will benefit from CRT. Furthermore, recent evidence suggests that patient response can be improved by placing the LV pacing lead in the most dyssynchronous or latest contracting segment. The overall goal of this project was to develop methods that allow for accurate assessment and display of regional mechanical dyssynchrony throughout the LV and at the site of the LV pacing lead. To accomplish this goal, we developed a method for quantifying regional dyssynchrony from standard short-axis cine magnetic resonance (MR) images. To assess the effects of LV lead placement, we developed a registration method that allows us to project the LV lead location from dual-plane fluoroscopy onto MR measurements of cardiac function. By applying these techniques in patients undergoing CRT, we were able to investigate the relationship between regional dyssynchrony, LV pacing lead location, and CRT response.