One important topic in medical image analysis is segmenting wall structures from different cardiac medical imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI). This task is typically done by radiologists either manually or semi-automatically, which is a very time-consuming process. To reduce the laborious human efforts, automatic methods have become popular in this research. In this thesis, features insensitive to data variations are explored to segment the ventricles from CT images and extract the left atrium from MR images. As applications, the segmentation results are used to facilitate cardiac disease analysis. Specifically, 1. An automatic method is proposed to extract the ventricles from CT images by integrating surface decomposition with contour evolution techniques. In particular, the ventricles are first identified on a surface extracted from patient-specific image data. Then, the contour evolution is employed to refine the identified ventricles. The proposed method is robust to variations of ventricle shapes, volume coverages, and image quality. 2. A variational region-growing method is proposed to segment the left atrium from MR images. Because of the localized property of this formulation, the proposed method is insensitive to data variabilities that are hard to handle by globalized methods. 3. In applications, a geometrical computational framework is proposed to estimate the myocardial mass at risk caused by stenoses. In addition, the segmentation of the left atrium is used to identify scars for MR images of post-ablation.

Thesis (Ph.D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Skrinjar, Oskar; Committee Member: Oshinski, John; Committee Member: Tannenbaum, Allen; Committee Member: Vela, Patricio; Committee Member: Yezzi, Anthony

Atrial fibrillation (AF), the most common type of cardiac arrythmia, has been a huge concern of public health. It affects more than 6 million people in the United States and over 33 million people worldwide. In the current standard of care, an electrogram and geometry map is generated by electroanatomic mapping (EAM) using a mapping catheter, to determine the origins of irregular heart rhythm. Followed by radiofrequency ablation (RFA) using ablation catheter, the targeted sites are ablated as lesions to change the electrical conduction pathway of abnormal electrogram, thus restoring the patients to normal sinus heart rhythm with minimally invasive procedure. However, a significant proportion of patients suffer from AF recurrences and requires repeated procedures, due to the lack of reliable methods to assess the cardiac structural substrates which are the potential maintaining mechanism of AF signals. In recent years, optical imaging modalities are developed to compensate this limitation, among which near-infrared spectroscopy (NIRS) is a catheter-based technique to enable direct, independent characterization of cardiac tissue pathology from spectrum morphology. In this thesis, we validate the capability of NIRS to generate map with repeatability and identify AF substrate to improve the efficacy of treatment. First, a near-infrared imaging spectroscopy was combined with an electromagnetic tracking modality, and the system was operated with high acquisition speed and real-time display to generate high-density map. Further, the robustness of NIRS optical parameters was assessed under blood mapping and various, large catheter-tissue contact angle, to simulate the dynamic circumstance of clinical procedures. A classification algorithm was introduced to predict lesion probability including both PBS and blood data, as well as to evaluate the mapping equivalence of blood and PBS. Next, the spatial resolution and the sampling density requirement of NIRS mapping method was characterized based on small gap, and the spectral properties of gap was assessed comparing to normal tissue and lesion by statistical analysis and machine learning. Lastly, we demonstrate the identification of human left atrial complex substrates using NIRS catheter with different source-detector-separations (SDSs), and reported the spectral features for the AF-related structures such as fibrosis and adipose. To summarize, the catheter-based NIRS technology is robust for in-vivo application and structural target localization, with the potential to enhance the recognition of underlying AF pathology and improve treatment efficacy.

The non-invasive cardiovascular imaging modalities, cardiovascular magnetic resonance (CMR) and multi-detector computer tomography (MDCT) are playing an increasing role in both clinical and research settings. CMR is a unique imaging modality due to unsurpassed contrast between soft tissue structures that is non-invasive, does not use ionising radiation and is able to provide high-resolution information about cardiac anatomy, function, flow, perfusion, viability and metabolism. It has provided the gold standard in imaging in congenital heart disease. Recent advances in this technology have led to images of high spatial and temporal resolution that has made the characterisation of atheroma possible. While currently spatial resolution still limits its ability to characterise atheroma in native human coronary arteries in living patients, CMR imaging of the coronary arteries has future potential with further technological and sequence advances. MDCT has been used in clinical settings to measure of the amount of calcification in the coronary arteries with “coronary artery calcium scoring” of the coronary tree a surrogate marker of atherosclerosis. MDCT has also become the gold standard for angiographic imaging in most arterial beds such as the carotid and peripheral vascular systems. In the coronary arteries in particular, there have been major advances in the accuracy of coronary MDCT angiography, particularly with regards to its negative predictive value, although excessive calcification and blooming artefacts still limit the diagnostic accuracy of the technique for assessing stenotic severity. In this thesis, our aims were to address some specific novel areas advancing the utility of these imaging modalities in two major areas of interest, namely congenital heart disease and atheroma imaging. Our first step was to validate the accuracy and reproducibility of CMR, the main imaging modality we utilised. To achieve this, we assessed MR imaging of cardiac volumes and function in a normal adult Australian population with a specific focus on the reproducibility of the technique. In confirming that this technique in our hands is both accurate and reproducible, we would then be in a position to be able to confidently use this technique in our future chapters. However, more than this, we sought to establish some normal ranges for left and right atrial and ventricular parameters in our local population. This would be crucial background information for us to be able to make comparisons with future studies in patients with congenital heart disease. Having established our technique and reference ranges, we would then explore the two specific issues in the ensuing two chapters using CMR in one area of congenital heart disease, atrial septal defect. Atrial septal defect is the most common congenital heart defect first diagnosed in adults. The traditional method of assessment of these patients and for suitability for ASD closure involves semiinvasive investigation with transoesophageal echocardiography (TOE) for measurement of the defect size and atrial septal margins. MRI assessment of patients prior to percutaneous device closure compared to TOE assessment would provide information on the accuracy of TOE assessment and provide information of the utility of cardiac MRI as an alternative to TOE for the work-up of these patients prior to ASD closure. In our third original research chapter, we utilised CMR to understand the effects of percutaneous ASD closure on cardiac chamber volumes. We achieved this by assessing with cardiac MRI pre-closure and post-closure atrial and ventricular cardiac volumes. Longstanding right heart dilatation in the setting of an ASD may lead to complications including right heart failure, pulmonary hypertension and arrhythmia. Closure of the ASD should reduce right heart volumes by removing left-to-right shunting and lead to normalisation of ventricular volumes. The assessment of atrial volume changes with ASD closure may be important in furthering our understanding in its contribution to arrhythmia. Having assessed the ability of CMR to assess both the ASD dimensions, and therefore suitability for percutaneous closure, as well as the effects of ASD closure on cardiac chamber size, we look in the final two original research chapters to move to another area of research development with these highresolution imaging technologies, atherosclerosis imaging. Two particular areas we wished to focus on included the potential of high-resolution MR imaging to monitor effects of HDL infusion on atherosclerosis, and secondly to explore mechanisms behind limitations in MDCT imaging of atherosclerosis, specifically calcification and blooming artifacts. For assessing the effects of HDL infusion on atherosclerosis, we utilised a cholesterol-fed rabbit model of atherosclerosis. The abdominal aorta of the rabbit is comparable in size to the human coronary artery. Previous work with the rabbit model of atherosclerosis and magnetic resonance imaging of the aortic wall has shown that it can provide information about atherosclerotic composition as well as provide serial data of the arterial wall. While high intensity lipid-lowering with statins remains the first line management of at risk individuals, modest manipulations of serum HDL levels are associated with a significant impact on cardiovascular risk. Thus, we assessed the effect of HDL infusion and atorvastatin in a rabbit model of using MRI aortic atherosclerosis as the endpoint. In our fifth and final original research chapter, we assessed the accuracy of quantification of atherosclerotic calcification with MDCT in the carotid arteries of patients undergoing carotid endarterectomy, and sought to identify algorithms or techniques that may improve quantification of calcification. This would potentially lead to an improvement in the ability of MDCT techniques to quantify stenotic severity in coronary arteries that were calcified. To achieve these we utilised MDCT in vivo and in comparison with carotid endarterectomy specimen micro-CT. Importantly, as part of this study, we undertook a thorough assessment of reproducibility of these techniques. Thus, in summary, we have been able to confirm the accuracy and reproducibility of CMR and MDCT in the areas of a specific congenital defect (ASD) and atherosclerosis imaging, and utilised these techniques to advance our understanding of these disease states. This thesis identifies strengths and weaknesses of these techniques that will allow us to more appropriately use them for future purposes in cardiovascular disease. Future work directly stemming from this thesis has already begun, and now looks to address issues of whether CMR and MDCT may provide complimentary information about atherosclerotic lesions that may benefit outcomes in certain conditions. Specifically the work in this thesis has led to studies commencing in carotid atherosclerosis and saphenous vein graft atherosclerosis and using these imaging techniques to potentially predict adverse future outcomes. / Thesis (Ph.D.) -- University of Adelaide, School of Medical Sciences, 2008

No description available.

The non-invasive cardiovascular imaging modalities, cardiovascular magnetic resonance (CMR) and multi-detector computer tomography (MDCT) are playing an increasing role in both clinical and research settings. CMR is a unique imaging modality due to unsurpassed contrast between soft tissue structures that is non-invasive, does not use ionising radiation and is able to provide high-resolution information about cardiac anatomy, function, flow, perfusion, viability and metabolism. It has provided the gold standard in imaging in congenital heart disease. Recent advances in this technology have led to images of high spatial and temporal resolution that has made the characterisation of atheroma possible. While currently spatial resolution still limits its ability to characterise atheroma in native human coronary arteries in living patients, CMR imaging of the coronary arteries has future potential with further technological and sequence advances. MDCT has been used in clinical settings to measure of the amount of calcification in the coronary arteries with “coronary artery calcium scoring” of the coronary tree a surrogate marker of atherosclerosis. MDCT has also become the gold standard for angiographic imaging in most arterial beds such as the carotid and peripheral vascular systems. In the coronary arteries in particular, there have been major advances in the accuracy of coronary MDCT angiography, particularly with regards to its negative predictive value, although excessive calcification and blooming artefacts still limit the diagnostic accuracy of the technique for assessing stenotic severity. In this thesis, our aims were to address some specific novel areas advancing the utility of these imaging modalities in two major areas of interest, namely congenital heart disease and atheroma imaging. Our first step was to validate the accuracy and reproducibility of CMR, the main imaging modality we utilised. To achieve this, we assessed MR imaging of cardiac volumes and function in a normal adult Australian population with a specific focus on the reproducibility of the technique. In confirming that this technique in our hands is both accurate and reproducible, we would then be in a position to be able to confidently use this technique in our future chapters. However, more than this, we sought to establish some normal ranges for left and right atrial and ventricular parameters in our local population. This would be crucial background information for us to be able to make comparisons with future studies in patients with congenital heart disease. Having established our technique and reference ranges, we would then explore the two specific issues in the ensuing two chapters using CMR in one area of congenital heart disease, atrial septal defect. Atrial septal defect is the most common congenital heart defect first diagnosed in adults. The traditional method of assessment of these patients and for suitability for ASD closure involves semiinvasive investigation with transoesophageal echocardiography (TOE) for measurement of the defect size and atrial septal margins. MRI assessment of patients prior to percutaneous device closure compared to TOE assessment would provide information on the accuracy of TOE assessment and provide information of the utility of cardiac MRI as an alternative to TOE for the work-up of these patients prior to ASD closure. In our third original research chapter, we utilised CMR to understand the effects of percutaneous ASD closure on cardiac chamber volumes. We achieved this by assessing with cardiac MRI pre-closure and post-closure atrial and ventricular cardiac volumes. Longstanding right heart dilatation in the setting of an ASD may lead to complications including right heart failure, pulmonary hypertension and arrhythmia. Closure of the ASD should reduce right heart volumes by removing left-to-right shunting and lead to normalisation of ventricular volumes. The assessment of atrial volume changes with ASD closure may be important in furthering our understanding in its contribution to arrhythmia. Having assessed the ability of CMR to assess both the ASD dimensions, and therefore suitability for percutaneous closure, as well as the effects of ASD closure on cardiac chamber size, we look in the final two original research chapters to move to another area of research development with these highresolution imaging technologies, atherosclerosis imaging. Two particular areas we wished to focus on included the potential of high-resolution MR imaging to monitor effects of HDL infusion on atherosclerosis, and secondly to explore mechanisms behind limitations in MDCT imaging of atherosclerosis, specifically calcification and blooming artifacts. For assessing the effects of HDL infusion on atherosclerosis, we utilised a cholesterol-fed rabbit model of atherosclerosis. The abdominal aorta of the rabbit is comparable in size to the human coronary artery. Previous work with the rabbit model of atherosclerosis and magnetic resonance imaging of the aortic wall has shown that it can provide information about atherosclerotic composition as well as provide serial data of the arterial wall. While high intensity lipid-lowering with statins remains the first line management of at risk individuals, modest manipulations of serum HDL levels are associated with a significant impact on cardiovascular risk. Thus, we assessed the effect of HDL infusion and atorvastatin in a rabbit model of using MRI aortic atherosclerosis as the endpoint. In our fifth and final original research chapter, we assessed the accuracy of quantification of atherosclerotic calcification with MDCT in the carotid arteries of patients undergoing carotid endarterectomy, and sought to identify algorithms or techniques that may improve quantification of calcification. This would potentially lead to an improvement in the ability of MDCT techniques to quantify stenotic severity in coronary arteries that were calcified. To achieve these we utilised MDCT in vivo and in comparison with carotid endarterectomy specimen micro-CT. Importantly, as part of this study, we undertook a thorough assessment of reproducibility of these techniques. Thus, in summary, we have been able to confirm the accuracy and reproducibility of CMR and MDCT in the areas of a specific congenital defect (ASD) and atherosclerosis imaging, and utilised these techniques to advance our understanding of these disease states. This thesis identifies strengths and weaknesses of these techniques that will allow us to more appropriately use them for future purposes in cardiovascular disease. Future work directly stemming from this thesis has already begun, and now looks to address issues of whether CMR and MDCT may provide complimentary information about atherosclerotic lesions that may benefit outcomes in certain conditions. Specifically the work in this thesis has led to studies commencing in carotid atherosclerosis and saphenous vein graft atherosclerosis and using these imaging techniques to potentially predict adverse future outcomes. / Thesis (Ph.D.) -- University of Adelaide, School of Medical Sciences, 2008

The study of myocardial motion is essential for understanding the normal heart function and developing new treatments for cardiovascular diseases. The goals of my PhD research is to develop new methods for cardiac deformation recovery from 3D magnetic resonance (MR) images. The main contribution of my work is that the proposed methods are guaranteed to generate exactly or nearly incompressible deformations. This is a desirable property since the myocardium has been shown to be close to incompressible. From the recovered deformation, one can directly compute a number of clinically useful parameters, including strains. The first method for 3D deformation recovery of the left ventricular wall (LV) from anatomical cine MRI is based on a deformable model that is incompressible. This method is not suitable for the deformation recovery of the biventricular wall. The second method is based on a 3D deformable model that is nearly incompressible. The model uses a matrix-valued radial basis function to represent divergence free displacement fields, which is a first order approximation of incompressibility. This representation allows for deformation modeling of arbitrary topologies with a relatively small number of parameters, which is suitable for representing the motion of the multi-chamber structure of the heart. The two methods have similar performance. A method to obtain a smooth and accurate surface of the myocardium wall is needed to illustrate the cardiac deformation recovery. I present a novel method for the generation of endocardial and epicardial surface meshes. The same algorithm is independently used to generate the surface meshes of the epicardium and endocardium of the four cardiac chambers. It provides smooth meshes despite the strong voxel anisotropy, which is not the case for the marching cubes algorithm. Phase velocity MRI is an acquisition technique that contains more information about the myocardial motion than cine MRI. I present a method to interpolate the velocity vector field in a phase velocity MRI sequence. The method uses an interpolation model that provides a continuous divergence free velocity vector field, which means that the corresponding deformation is incompressible.

Thesis (PhD)-- Stellenbosch University, 2013. / ENGLISH ABSTRACT: This dissertation considers automatic segmentation of the left cardiac ventricle in short axis magnetic resonance images. The presence of papillary muscles near the endocardium border makes simple threshold based segmentation difficult. The endo- and epicardium are modelled as two series of radii which are inter-related using features describing shape and motion. Image features are derived from edge information from human annotated images. The features are combined within a Conditional Random Field (CRF) – a discriminatively trained probabilistic model. Loopy belief propagation is used to infer segmentations when an unsegmented video sequence is given. Powell’s method is applied to find CRF parameters by minimising the difference between ground truth annotations and the inferred contours. We also describe how the endocardium centre points are calculated from a single human-provided centre point in the first frame, through minimisation of frame alignment error. We present and analyse the results of segmentation. The algorithm exhibits robustness against inclusion of the papillary muscles by integrating shape and motion information. Possible future improvements are identified. / AFRIKAANSE OPSOMMING: Hierdie proefskrif bespreek die outomatiese segmentasie van die linkerhartkamer in kortas snit magnetiese resonansie beelde. Die teenwoordigheid van die papillêre spiere naby die endokardium grens maak eenvoudige drumpel gebaseerde segmentering moeilik. Die endo- en epikardium word gemodelleer as twee reekse van die radiusse wat beperk word deur eienskappe wat vorm en beweging beskryf. Beeld eienskappe word afgelei van die rand inligting van mens-geannoteerde beelde. Die funksies word gekombineer binne ’n CRF (Conditional Random Field) – ’n diskriminatief afgerigte waarskynlikheidsverdeling. “Loopy belief propagation” word gebruik om segmentasies af te lei wanneer ’n ongesegmenteerde video verskaf word. Powell se metode word toegepas om CRF parameters te vind deur die minimering van die verskil tussen mens geannoteerde segmentasies en die afgeleide kontoere. Ons beskryf ook hoe die endokardium se middelpunte bereken word vanaf ’n enkele mens-verskafte middelpunt in die eerste raam, deur die minimering van ’n raambelyningsfout. Ons analiseer die resultate van segmentering. Die algoritme vertoon robuustheid teen die insluiting van die papillêre spiere deur die integrasie van inligting oor die vorm en die beweging. Moontlike toekomstige verbeterings word geïdentifiseer.

Zu den wichtigsten Werkzeugen in der medizinischen Diagnostik gehören bildgebende Verfahren, wie die Magnetresonanztomographie. Ein weiteres diagnostisches Hilfsmittel ist die Palpation, die es erlaubt, Veränderungen oberflächennaher Organe qualitativ zu erfassen. Die Magnetresonanzelastographie (MRE) stellt eine Kombination dieser Techniken dar. Das Prinzip der MRE besteht darin Gewebedeformationen aufgrund extern induzierter Scherwellen mittels bewegungssensitiver MR-Bildgebung darzustellen und über die Art der Deformation auf die Elastizität des Gewebes zu schließen. Einen großen Anteil schwerwiegender Erkrankungen bilden Störungen des Herz-Kreislaufsystems. Das Ziel dieser Arbeit war es, eine Methode zu entwickeln, die in-vivo MRE am menschlichen Herzen ermöglicht. Die Weiterentwicklung der mechanischen Anregungseinheit ergab mit Einführung eines Audio-Lautsprechers das nötige Instrument, Vibrationen in innere Organe zu übertragen. Der entscheidende Faktor bei der Herz-MRE war die Geschwindigkeit der Aufnahme, die zur Entwicklung der ''fraktionierten MRE'' führte. Die Basis waren schnelle Herzbildgebungstechniken, wie die balancierte Steady-State- (bSSFP) und Spoiled Steady-State-Technik (SPGRE). Die Einführung eines unbalancierten Phasenpräparationsgradienten in der bSSFP-Aufnahmetechnik lieferte ein verbessertes SNR und zusammen mit der SPGRE-MRE-Aufnahmetechnik ließen sich damit MRE-Studien auch am menschlichen Herzen durchführen. Es gelang erstmals, extern induzierte mechanische Schwingungen in das Herz zu koppeln und mittels fraktionierter MRE mit hoher zeitlicher Auflösung zu detektieren. Die in 6 gesunden Probanden beobachtete Modulation der Scherwellenamplituden innerhalb des Myokards korrelierte sehr gut mit den Kontraktionszuständen des Herzens. Die entwickelten Techniken und Methoden sind ein Schritt hin zur routinemäßigen klinischen Anwendung der MRE am Herzen und deuten auf ein hohes Potential im Bereich der Diagnostik kardialer Erkrankungen hin. / Imaging techniques, including magnetic resonance imaging, belong to the most important tools in modern medical diagnostics. Another diagnostic aid is palpation, which is suitable for the qualitative characterization of pathological changes in organs near the surface. Magnetic resonance elastography (MRE) is a combination of these techniques. In principle, MRE uses motion-sensitive MR-imaging to depict tissue deformation caused by externally induced shear waves. The type of deformation supply useful information about the elasticity of the tissue. Cardiac disorders are among the most common diseases. The goal of this study was to develop a method of applying in-vivo MRE to the human heart. The development of the mechanical stimulus, ultimately resulting in the introduction of an audio speaker as the source of vibration, provided the necessary means to introduce vibrations into inner organs. A crucial factor in applying MRE to the heart is the speed of the recording, which led to the development of "fractional MRE". The currently conventional fast heart imaging techniques were used as a starting point. The use of an unbalanced phase preparation gradient in the balanced steady-state imaging technique resulted in an improved phase-to-noise ratio. Along with the spoiled steady-state MRE imaging technique, initial MRE-studies on the human heart were performed. For the first time, externally induced mechanical vibrations were successfully introduced into the heart and were detected using fractional MRE with a high temporal resolution. The modulation of the shear wave amplitudes observed in the myocard of 6 healthy subjects correlated with the phases of the cardiac cycle. The techniques and methods developed here are a step toward routine clinical application of MRE of the heart and indicate high potential in the area of early diagnosis of cardiac disease.

Indiana University-Purdue University Indianapolis (IUPUI) / The sympathetic nervous system strongly modulates the contractile and electrical function of the heart. The anatomical underpinnings that enable a spatially and temporally coordinated dissemination of sympathetic signals within the cardiac tissue are only incompletely characterized. In this work we took the first step of unraveling the in situ 3D microarchitecture of the cardiac sympathetic nervous system. Using a combination of two-photon excitation fluorescence microscopy and computer-assisted image analyses, we reconstructed the sympathetic network in a portion of the left ventricular epicardium from adult transgenic mice expressing a fluorescent reporter protein in all peripheral sympathetic neurons. The reconstruction revealed several organizational principles of the local sympathetic tree that synergize to enable a coordinated and efficient signal transfer to the target tissue. First, synaptic boutons are aligned with high density along much of axon-cell contacts. Second, axon segments are oriented parallel to the main, i.e., longitudinal, axes of their apposed cardiomyocytes, optimizing the frequency of transmitter release sites per axon/per cardiomyocyte. Third, the local network was partitioned into branched and/or looped sub-trees which extended both radially and tangentially through the image volume. Fourth, sub-trees arrange to not much overlap, giving rise to multiple annexed innervation domains of variable complexity and configuration. The sympathetic network in the epicardial border zone of a chronic myocardial infarction was observed to undergo substantive remodeling, which included almost complete loss of fibers at depths >10 µm from the surface, spatially heterogeneous gain of axons, irregularly shaped synaptic boutons, and formation of axonal plexuses composed of nested loops of variable length. In conclusion, we provide, to the best of our knowledge, the first in situ 3D reconstruction of the local cardiac sympathetic network in normal and injured mammalian myocardium. Mapping the sympathetic network connectivity will aid in elucidating its role in sympathetic signal transmisson and processing.