Dissertations / Theses on the topic 'Computed tomography ; Cardiac imaging ; Cardiology'

Throughout the development of computed tomographic (CT) imaging the challenges of capturing the heart, with its perpetual, vigorous motion, and in particular the tiny detail within the coronary arteries, has driven technological progress. Today, CT is a widely used and rapidly growing modality for the investigation of coronary artery disease, as well as other cardiac pathology. However, limitations remain and particular patient groups present a significant challenge to the CT operator. This thesis adds new knowledge to the assessment of these difficult-to-image patients. It considers patients with artefact from coronary artery calcification or stents, examining the remarkable diagnostic performance of high definition scanning, as well as material subtraction techniques using dual energy CT, alongside ways in which current technology might be revisited and refined with the use of alternative image reconstruction methods. Patients with challenging heart rate or rhythm abnormalities are considered in three studies; how to achieve diagnostic image quality in atrial fibrillation, the safety of an aggressive approach to intravenous beta-blocker use prior to coronary imaging, and the development of patient information to address anxiety as a source of tachycardia and motion artefact. Finally, the novel application of a single source, dual energy CT scanner to additional cardiac information is considered, with studies of myocardial perfusion CT and delayed iodine enhancement imaging, to identify ways in which non-coronary imaging might be exploited to more thoroughly evaluate a patient’s coronary artery status. These findings are presented in the context of developing technology and together offer a range of potential options for operators of cardiac CT when faced with a difficult-to-image patient.

Background Cardiac Computed Tomography Angiography (CCTA) is becoming increasingly useful in the work‐up of coronary artery disease (CAD). Several potential methods for increasing the diagnostic yield of cardiac CT are available. Purpose Study I: To investigate whether the use of a 2‐D, non‐linear adaptive noise reduction filter can improve CCTA image quality. Study II: To evaluate the variation in adenosine stress dynamic CT perfusion (CTP) blood flow as compared to stress 99mTc SPECT. Secondly, to compare the perfusion results from manual and automatic myocardial CTP segmentation. Study III: To evaluate the accuracy of non‐invasive, CCTA‐derived Fractional Flow Reserve (cFFR). Study IV: To evaluate the prognostic value of CCTA in terms of major adverse cardiac events (MACE). Materials and methods Study I: Single images from 36 consecutive CCTA exams performed with two different dose levels were used. Image quality in full dose, low‐dose and noise‐reduced low‐dose images was graded using visual grading analysis. Image noise was measured. Study II: CTP and SPECT were performed in 17 patients, and the variation in per AHA‐segment blood flow was evaluated and compared. CTP results from manual and automated image segmentation were compared. Study III: CCTA datasets from 21 patients were processed using cFFR software and the results compared to the corresponding invasively measured FFR (invFFR). Study IV: 1205 consecutive patients with chest pain of unknown origin underwent CCTA. Baseline data and data on subsequent MACE were retrieved from relevant registries. Survival, hazard ratios and the three‐year incidence of cardiac events and readmissions were calculated. Results Study I: There was significant improvement in perceived image quality for all criteria when the filter was applied, and a significant decrease in image noise. Study II: The correlation coefficients for CTP vs. SPECT were 0.38 and 0.41 (p<0.001, for manual and automated segmentation respectively. Mean per patient CTP blood flow in normal segments varied between 94‐183 ml/100 ml tissue/min for manual segmentation, and 104‐196 ml/100 ml tissue/min for automated segmentation. The Spearman rank correlation coefficient for manual vs. automated segmentation CTP was ρ = 0.88 (p<0.001) and the Intraclass Correlation Coefficient (ICC) was 0.93 (p<0.001). Study III: The Spearman rank correlation coefficient for cFFR vs. invFFR was ρ = 0.77 (p<0.001) and the ICC was 0.73 (p<0.001). Sensitivity, specificity, positive predictive value and negative predictive value for significant stenosis (FFR<0.80, per vessel) were 0.83, 0.76, 0.56 and 0.93 respectively. Study IV: The hazard ratio for non‐obstructive CAD vs. normal coronary arteries was 5.13 (95% C.I 1.03‐25.43, p<0.05), and 151.40 (95% C.I 37.03‐619.08, p<0.001) for obstructive CAD vs. normal coronary arteries. The three‐year incidence of MACE was 1.1% for patients with normal vessels on CCTA, 2.5% for patients with non‐obstructive CAD and 42.7% for patients with obstructive CAD (p<0.001). Conclusions: Study I: Image quality and noise levels of low dose images were significantly improved with the filter, even though the improvement was small compared to the image quality of the corresponding diastolic full‐dose images. Study II: Correlation between dynamic CTP and SPECT was positive but weak. There were large variations in CTP blood flow in normal segments on SPECT, rendering the definition of an absolute cut‐off value for normal vs. ischemic myocardium difficult. Manual and automatic segmentation were equally useful. Study III: The correlation between cFFR and invFFR was good, indicating that noninvasively estimated cFFR performs on a similar level as invasively measure FFR. Study IV: The long‐term risk for MACE was very low in patients without obstructive CAD on CCTA, though there seemed to be a substantial increase in the risk for MACE even in patients with non‐obstructive CAD as compared to normal coronary arteries. In addition, even patients with normal coronary arteries or non‐obstructive CAD continued to have a substantial number of readmissions for chest pain or angina pectoris.

Cardiovascular diseases are the largest mortality factor globally, and early diagnosis is essential for a proper medical response. Cardiac computed tomography can be used to acquire images for their diagnosis, but without radiation dose reduction the radiation emitted to the patient becomes a significant risk factor. By reducing the dose, the image quality is often compromised, and determining a diagnosis becomes difficult. This project proposes image quality enhancement with deep learning. A cycle-consistent generative adversarial neural network was fed low- and high-quality images with the purpose to learn to translate between them. By using a cycle-consistency cost it was possible to train the network without paired data. With this method, a low-quality image acquired from a computed tomography scan with dose reduction could be enhanced in post processing. The results were mixed but showed an increase of ventricular contrast and artifact mitigation. The technique comes with several problems that are yet to be solved, such as structure alterations, but it shows promise for continued development.

ECG-gated multidetector computed tomography (MDCT) is an imaging modality widely utilized for the evaluation of cardiac pathology by physicians. However, there has been little research of cardiac MDCT imaging in veterinary patients. Presently, ECG-gating is an upgrade for MDCT, which few veterinary institutions currently possess. The purpose of this study was to compare image quality between a 16 non ECG-gated and 64 ECG-gated MDCT for clinically important cardiac anatomy in dogs. In a crossover trial, six dogs were scanned using 16 non ECG-gated and 64 ECG-gated MDCT. A standardized anesthetic protocol, designed to induce bradycardia (mean HR 45 bpm ± 12.6) was used. Five post-contrast sequential scans through the heart were performed for each patient when utilizing the 16 non ECG-gated MDCT, in attempt to obtain a motion free series of images of the heart. For each scan, assessment of cardiac morphology was performed by evaluating a group of 21 cardiac structures, using a 3-point scale. Each of the images were scored as 0 (motion present, scan non-diagnostic), 1 (motion present, scan diagnostic), and 2 (no motion, therefore diagnostic scan of high quality). Quality scores (QS) from all scans within a dog (30 scans total) were assigned for each cardiac structure. QS from the six ECG-gated MDCT scans were of high diagnostic quality, generating diagnostic images for all of the 21 cardiac structures evaluated for each of the 6 scans. Individual non ECG-gated scans were of variable quality, primarily generating QS of 1 or 2. A complete set of diagnostic images for all 21 structures was not achieved from an individual scan. Minimum number of non ECG-gated scans to identify a single structure was calculated, and ranged from 1-2 scans for all structures. Cumulative number of sequential non ECG-gated scans needed to achieve images of all cardiac structures was calculated and determined to be 5. A 16 non ECG-gated MDCT scanner can produce cardiac images that are similar in quality, to those of 64 ECG-gated MDCT. Cardiac motion negatively impacts image quality in studies acquired without ECG-gating. However, this can be overcome by performing multiple sequential scans through the heart.
Master of Science

BACKGROUND Aortic stenosis is a common and potentially fatal condition in which fibro-calcific changes within the valve leaflets lead to the obstruction of blood flow. Severe symptomatic stenosis is an indication for aortic valve replacement and timely referral is essential to prevent adverse clinical events. Calcification is believed to represent the central process driving disease progression. 18F-Fluoride positron emission tomography computed tomography (PET-CT) and CT aortic valve calcium scoring (CT-AVC) quantify calcification activity and burden respectively. The overarching aim of this thesis was to evaluate the applications of these techniques to the study and management of aortic stenosis. METHODS AND RESULTS REPRODUCIBILITY The scan-rescan reproducibility of 18F-fluoride PET-CT and CT-AVC were investigated in 15 patients with mild, moderate and severe aortic stenosis who underwent repeated 18F-fluoride PET-CT scans 3.9±3.3 weeks apart. Modified techniques enhanced image quality and facilitated clear localization of calcification activity. Percentage error was reduced from ±63% to ±10% (tissue-to-background ratio most-diseased segment (MDS) mean of 1.55, bias -0.05, limits of agreement - 0·20 to +0·11). Excellent scan-rescan reproducibility was also observed for CT-AVC scoring (mean of differences 2% [limits of agreement, 16 to -12%]). AORTIC VALVE CALCIUM SCORE: SINGLE CENTRE STUDY Sex-specific CT-AVC thresholds (2065 in men and 1271 in women) have been proposed as a flow-independent technique for diagnosing severe aortic stenosis. In a prospective cohort study, the impact of CT-AVC scores upon echocardiographic measures of severity, disease progression and aortic valve replacement (AVR)/death were examined. Volunteers (20 controls, 20 with aortic sclerosis, 25 with mild, 33 with moderate and 23 with severe aortic stenosis) underwent CT-AVC and echocardiography at baseline and again at either 1 or 2-year time-points. Women required less calcification than men for the same degree of stenosis (p < 0.001). Baseline CT-AVC measurements appeared to provide the best prediction of subsequent disease progression. After adjustment for age, sex, peak aortic jet velocity (Vmax) ≥ 4m/s and aortic valve area (AVA) < 1 cm2, the published CT-AVC thresholds were the only independent predictor of AVR/death (hazard ratio = 6.39, 95% confidence intervals, 2.90-14.05, p < 0.001). AORTIC VALVE CALCIUM SCORE: MULTICENTRE STUDY CT-AVC thresholds were next examined in an international multicenter registry incorporating a wide range of patient populations, scanner vendors and analysis platforms. Eight centres contributed data from 918 patients (age 77±10, 60% male, Vmax 3.88±0.90 m/s) who had undergone ECG-gated CT within 3 months of echocardiography. Of these 708 (77%) had concordant echocardiographic assessments, in whom our own optimum sex-specific CT-AVC thresholds (women 1377, men 2062 AU) were nearly identical to those previously published. These thresholds provided excellent discrimination for severe stenosis (c-statistic: women 0.92, men 0.88) and independently predicted AVR and death after adjustment for age, sex, Vmax ≥4 m/s and AVA < 1 cm2 (hazards ratio, 3.02 [95% confidence intervals, 1.83-4.99], p < 0.001). In patients with discordant echocardiographic assessments (n=210), CT-AVC thresholds predicted an adverse prognosis. BICUSPID AORTIC VALVES Within the multicentre study, higher continuity-derived estimates of aortic valve area were observed in patients with bicuspid valves (n=68, 1.07±0.35 cm) compared to those with tri-leaflet valves (0.89±0.36 cm p < 0.001,). This was despite no differences in measurements of Vmax (p=0.152), or CT-AVC scores (p=0.313). The accuracy of AVA measurments in bicuspid valves was therefore tested against alternative markers of disease severity. AVA measurements in bicuspid valves demonstrated extremely weak associations with CT-AVC scores (r2=0.08, p=0.02) and failed to correlate with downstream markers of disease severity in the valve and myocardium and against clinical outcomes. AVA measurements in bicuspid patients also failed to independently predict AVR/death after adjustment for Vmax ≥4 m/s, age and gender. In this population CT-AVC thresholds (women 1377, men 2062 AU) again provided excellent discrimination for severe stenosis. CONCLUSIONS Optimised 18F-fluoride PET-CT scans quantify and localise calcification activity, consolidating its potential as a biomarker or end-point in clinical trials of novel therapies. CT calcium scoring of aortic valves is a reproducible technique, which provides diagnostic clarity in addition to powerful prediction of disease progression and adverse clinical events.

Las enfermedades cardiovasculares (ECVs) son la principal causa de mortalidad en el mundo Occidental. El interés de prevenir y tratar las ECVs ha desencadenado un rápido desarrollo de los sistemas de adquisición de imágenes médicas. Por este motivo, la cantidad de datos de imagen recolectados en las instituciones de salud se ha incrementado considerablemente. Este hecho ha aumentado la necesidad de herramientas automatizadas para dar soporte al diagnóstico, mediante una interpretación de imagen confiable y reproducible. La tarea de interpretación requiere traducir los datos crudos de imagen en parámetros cuantitativos, los cuales son considerados relevantes para clasificar la condición cardiaca de un paciente. Para realizar tal tarea, los métodos basados en modelos estadísticos de forma han recibido favoritismo dada la naturaleza tridimensional (o 3D+t) de las imágenes cardiovasculares. Deformando el modelo estadístico de forma a la imagen de un paciente, el corazón puede analizarse de manera integral. Actualmente, el campo de las imágenes cardiovasculares esta constituido por diferentes modalidades. Cada modalidad explota diferentes fenómenos físicos, lo cual nos permite observar el órgano cardiaco desde diferentes ángulos. El personal clínico recopila todas estas piezas de información y las ensambla mentalmente en un modelo integral. Este modelo integral incluye información anatómica y funcional que muestra un cuadro completo del corazón del paciente. Es de alto interés transformar este modelo mental en un modelo computacional capaz de integrar la información de manera global. La generación de un modelo como tal no es simplemente un reto de visualización. Requiere una metodología capaz de extraer los parámetros cuantitativos relevantes basados en los mismos principios técnicos. Esto nos asegura que las mediciones se pueden comparar directamente. Tal metodología debe ser capaz de: 1) segmentar con precisión las cavidades cardiacas a partir de datos multimodales, 2) proporcionar un marco de referencia único para integrar múltiples fuentes de información, y 3) asistir la clasificación de la condición cardiaca del paciente. Esta tesis se basa en que los modelos estadísticos de forma, y en particular los Modelos Activos de Forma, son un método robusto y preciso con el potencial de incluir todos estos requerimientos. Para procesar múltiples modalidades de imagen, separamos la información estadística de forma de la información de apariencia. Obtenemos la información estadística de forma a partir de una modalidad de alta resolución y aprendemos la apariencia simulando la física de adquisición de otras modalidades. Las contribuciones de esta tesis pueden ser resumidas así: 1) un método genérico para construir automáticamente modelos de intensidad para los Modelos Activos de Forma simulando la física de adquisición de la modalidad en cuestión, 2) la primera extensión de un simulador de Resonancia Magnética Nuclear diseñado para producir estudios cardiacos realistas, y 3) un método novedoso para el entrenamiento automático de modelos de intensidad y de fiabilidad aplicado a estudios cardiacos de Resonancia Magnética Nuclear. Cada una de estas contribuciones representa un artículo publicado o enviado a una revista técnica internacional.
Cardiovascular diseases (CVDs) are the major cause of death in the Western world. The desire to prevent and treat CVDs has triggered a rapid development of medical imaging systems. As a consequence, the amount of imaging data collected in health care institutions has increased considerably. This fact has raised the need for automated analysis tools to support diagnosis with reliable and reproducible image interpretation. The interpretation task requires to translate raw imaging data into quantitative parameters, which are considered relevant to classify the patient’s cardiac condition. To achieve this task, statistical shape model approaches have found favoritism given the 3D (or 3D+t) nature of cardiovascular imaging datasets. By deforming the statistical shape model to image data from a patient, the heart can be analyzed in a more holistic way. Currently, the field of cardiovascular imaging is constituted by different modalities. Each modality exploits distinct physical phenomena, which allows us to observe the cardiac organ from different angles. Clinicians collect all these pieces of information to form an integrated mental model. The mental model includes anatomical and functional information to display a full picture of the patient’s heart. It is highly desirable to transform this mental model into a computational model able to integrate the information in a comprehensive manner. Generating such a model is not simply a visualization challenge. It requires having a methodology able to extract relevant quantitative parameters by applying the same principle. This assures that the measurements are directly comparable. Such a methodology should be able to: 1) accurately segment the cardiac cavities from multimodal datasets, 2) provide a unified frame of reference to integrate multiple information sources, and 3) aid the classification of a patient’s cardiac condition. This thesis builds upon the idea that statistical shape models, in particular Active Shape Models, are a robust and accurate approach with the potential to incorporate all these requirements. In order to handle multiple image modalities, we separate the statistical shape information from the appearance information. We obtain the statistical shape information from a high resolution modality and include the appearance information by simulating the physics of acquisition of other modalities. The contributions of this thesis can be summarized as: 1) a generic method to automatically construct intensity models for Active Shape Models based on simulating the physics of acquisition of the given imaging modality, 2) the first extension of a Magnetic Resonance Imaging (MRI) simulator tailored to produce realistic cardiac images, and 3) a novel automatic intensity model and reliability training strategy applied to cardiac MRI studies. Each of these contributions represents an article published or submitted to a peer-review archival journal.

I NTRODUÇÃO: Na cintilografia de perfusão do miocárdio, contrariamente ao desenvolvimento tecnológico ao longo do tempo, manteve-se desde o princípio a mesma forma, por vezes desconfortável ou mesmo inviável, de posicionamento na aquisição das imagens: decúbito dorsal com os braços elevados. OBJETIVOS: O objetivo da pesquisa foi o de avaliar se na aquisição da cintilografia de perfusão do miocárdio o posicionamento com os membros superiores abaixados(T) apresenta os mesmos resultados que os obtidos com os membros superiores elevados(C), tanto em termos de alterações perfusionais como em relação aos parâmetros funcionais. . MÉTODOS: Foram estudados 120 pacientes, 83 (69%) do sexo masculino, com idade 59,4 ± 11,4 anos e peso 72,8 ± 14 kg. Foi realizada primeiramente a aquisição C e na seqüência a aquisição T, utilizando protocolo de um dia de repouso-estresse (dose 370 MBq e 1110 MBq de sestamibi-99mTc), com estresse sincronizado ao eletrocardiograma (gated- SPECT). Três modelos distintos de equipamentos de dois detectores foram utilizados no estudo. Os estudos foram processados usando reconstrução iterativa (OSEM). Na interpretação foi utilizado o modelo de segmentação miocárdica de 17 segmentos, pontuados de 0 a 4 (normal a ausente) segundo o grau de captação. Baseada na análise de um observador dentre um grupo de sete, foi realizada a comparação da totalidade dos segmentos e também segmento a segmento das aquisições C e T nas etapas de estresse e de repouso. Foi comparada também a somatória das pontuações das etapas de estresse(SSS) e repouso(SRS). Os pacientes foram divididos segundo dois critérios: primeiramente, normais (SSS=0) e anormais (SSS>1) e posteriormente, em de baixo risco (SSS3) e de maior risco (SSS>3) A comparação dos parâmetros funcionais de fração de ejeção do ventrículo esquerdo (FEVE), volume diastólico final (VDF), volume sistólico final(VSF), somatória dos escores de motilidade (SMS) e somatória dos escores de espessamento (STS) foi baseada nos valores obtidos de modo automático utilizando o programa computacional Quantitative Gated SPECT. RESULTADOS: No modo C, 23,3% dos pts referiram algum grau de dor no membro superior ou no ombro e no modo T 5 %. Houve diferença significativa (p<0,05) na comparação dos 2040 segmentos miocárdicos entre C e T, ao estresse e ao repouso. Nos 63 pts normais não houve diferença significativa dos escores entre C e T ao estresse e ao repouso. O mesmo foi observado nos 80 pts de baixo risco. Nos 57 pts anormais houve boa correlação entre os valores de SSS (Rho=0,93, p=0,0001) e SRS (Rho=0,93, p=0,0001), mas com valores não equivalentes (p<0,05), sendo a média dos valores de SSS (9,28±8,10) e SRS (7,89±7,34) da aquisição T inferiores à média dos valores de SSS (10,07±7,71) e SRS (8,46±7,35) encontrada na aquisição C. Nos 40 pts de risco, houve boa correlação dos valores de SSS(Rho=0,95, p=0,0001) e SRS(Rho=0,96 p=0,0001), porém a média dos valores de SSS (12,53±7,54) e SRS (10,60±7,08) da aquisição T foi significantemente (p<0,05) inferior à média dos valores de SSS (13,43±6,81) e SRS(11,33±6,97) encontrados na aquisição C. Os parâmetros funcionais apresentaram boa correlação entre C e T, exceto o VDF(p=0,0001). CONCLUSÕES: Os modos C e T apresentam boa correlação dos parâmetros funcionais, exceto o VDF. Embora a aquisição T tenha sido um pouco mais confortável e tenha demonstrado boa correlação nos achados perfusionais, em pts anormais e de maior risco há tendência à subestimação dos defeitos perfusionais. Considerando a importância da medida precisa da intensidade e extensão das alterações perfusionais em termos de avaliação prognostica e decisão terapêutica, a aquisição com membro superior abaixado deve ser evitada. Em situações onde a aquisição com membros superiores abaixados é a única alternativa, os resultados da cintilografia devem ser interpretados com cuidado, especialmente em pacientes anormais e de maior risco, onde as alterações podem estar subestimadas
Contrary to the advances in imaging technology for nuclear cardiology applications, we keep using the same often uncomfortable and sometimes impracticable patient position- supine with arms raised above the head (C). We tested another position modality: supine with arms down at the sides of the trunk (T). The purpose of this study was to verify if the functional and perfusion results of the acquisition T are equivalent to those of the acquisition mode C. We studied 120 patients (pts), 83 (69%) male, aged 59.4±11.4 years and weighting 72.8±14 kg. We performed a one-day protocol (rest gated/stress), using 99mTc-sestamibi (370 MBq and 1110 MBq). In both times (rest and stress), we first performed acquisition in C and in sequence the acquisition in T. The studies were performed in three types of dual detector SPECT systems. T mode was executed successfully in all pts. Images were processed by the iterative reconstruction method (OSEM). Each study was independently interpreted by one nuclear medicine specialist from a group of seven physicians using the 17-segment model. The segments were scored using a 5-point model ranging from 0 (normal uptake) to 4 (uptake absent). The total score of the left ventricle at stress is referred to as the summed stress score (SSS) and at rest as the summed rest score (SRS). The patients were categorized in subgroups by two criteria: normal (SSS=0) or abnormal (SSS1) and low risk (SSS3) or risk (SSS>3). The values of the functional parameters of left ventricular ejection fraction (LVEF), end diastolic volume (EDV), end systolic volume (ESV), stress motion score (SMS) and stress thickening score (STS) for both C and T were automatically obtained by the quantitative gated SPECT (QGS) program and results were compared. Shoulder and/or back pain occurred in 23.3% of C patients and in 5% of T patients. There was no agreement between the 2040 segmental scores of both rest and stress in C and T modes (p<0.05). No significant differences between C and T were found for SSS and SRS in the 63 normal individuals nor in the 80 low risk patients. Good correlation between C and T was found for SSS (Rho=0.95, p=0.0001) and SRS (Rho=0.96 p=0.0001) in the 57 abnormal pts, but the mean SSS (9.28±8.10) and SRS (7.89±7.34) values of T were significantly lower (p<0.05) than the mean for SSS (10.07±7,71) and SRS(8.46±7,35) of C mode. Similar patterns were observed in the 40 risk pts; good correlation was found between C and T modes for SSS (Rho=0.95, p=0.0001) and SRS (Rho=0.96 p=0.0001), but the mean SSS (12.53±7.54) and SRS (10.60±7.08) values of T were significantly lower (p<0.05) than SSS (13.43±6.81) and SRS (11.33±6.97) of C mode. Good correlation between C and T was found for all functional parameters, except for EDV (p=0.0001). Although T mode appears to be more comfortable and presented a good correlation between SSS and SRS values, in abnormal and risk pts, the extent and severity of defects can be underestimated. Considering the important therapeutic and prognostic implications of an accurate perfusion measurement, the cardiac SPECT acquisition with the arms in down position should be avoided. When arms-down acquisition is the only alternative, scintigraphy results must be carefully interpreted, especially in abnormal scans whereas the amount of perfusion abnormalities can be underestimated.

Applying mathematical models to dynamic contrast enhanced MRI (DCE MRI) data to perform quantitative tracer kinetic analysis enables the estimation of tissue characteristics such as vascular permeability and the fractional volume of plasma in a tissue. However, it is unclear to what extent modeling assumptions, particularly regarding water exchange between tissue compartments, impacts parameter estimates derived from clinical DCE MRI data. In this work, a new model is developed which includes water exchange effects, termed the water exchange modified two compartment exchange model (WX-2CXM). Two boundaries of this model (the fast and no exchange limits) were used to analyse a clinical DCE MRI bladder cancer dataset. Comparisons with DCE CT, which is not affected by water exchange, suggested that water exchange may have affected estimates of vp, the fractional volume of plasma. Further investigation and simulations led to the development of a DCE MRI protocol which was sensitised to water exchange, in order to further evaluate the water exchange effects found in the bladder cancer dataset. This protocol was tested by imaging the parotid glands in eight healthy volunteers, and confirmed evidence of water exchange effects on vp, as well as flow Fp and the fractional volume of extravascular extracellular space ve. This protocol also enabled preliminary estimates of the water residence times in parotid tissue, however, these estimates had a large variability and require further validation. The work presented in this thesis suggests that, although water exchange effects do not have a large effect on clinical data, the effect is measurable, and may lead to the ability to estimate of tissue water residence times. Results do not support a change in the current practise of neglecting water exchange effects in clinical DCEMRI acquisitions.

This thesis presents three original and complementary approaches to enhance the quality of Statistical Shape Models (SSMs), that improve the accuracy of medical image segmentation in challenging applications. First, we enhance the statistical richness of SSMs by developing a technique capable of merging the shape representations and statistical properties of several pre-existing models with no original or additional raw data. Second, we enhance the geometrical quality of SSMs by developing a framework for modeling simultaneously both global and local characteristics of highly complex and/or multi-part anatomical shapes. Last, we improve the specificity of SSMs for specific subjects by integrating individual-specific non-imaging metadata such as demographic, clinical and behavioral variables into the SSM construction and image segmentation tasks. These techniques are demonstrated and validated by considering various imaging modalities such as magnetic resonance imaging (MRI) and computed tomography (CT), and different complex anatomies, including the human heart, brain and spine.
Esta tesis presenta tres propuestas originales y complementarias para mejorar la calidad de los modelos estadísticos de formas (SSMs) que mejoran la precisión de la segmentación de la imagen médica en aplicaciones difíciles. Proponemos, primero, mejorar la riqueza estadística de los SSMs por medio de una técnica para unir la representación de forma y las propiedades estadísticas de muchos modelos pre-existentes sin observaciones adicionales. Segundo, mejorar la representacion geométrica de los SSMs modelando simultáneamente las características globales y locales del objecto o de multiples anatomias. Por último, mejorar la especificidad de los SSMs mediante la integración de metadatos del paciente no derivados de la imagen, tales como, variables demográficas, conductuales y de entorno clínico, en la construcción de los modelos. Estas técnicas son demostradas y validadas en imágenes de resonancia magnética (MRI) y tomografía computarizada (CT) y en anatomias como el corazón, el cerebro y la espina dorsal humanos.

Les artéfacts métalliques entraînent un épaississement artéfactuel de la paroi des tuteurs en tomodensitométrie (TDM) avec réduction apparente de leur lumière. Cette étude transversale prospective, devis mesures répétées et observateurs avec méthode en aveugle, chez 24 patients consécutifs/71 tuteurs coronariens a pour objectif de comparer l’épaisseur de paroi des tuteurs en TDM après reconstruction par un algorithme avec renforcement des bords et un algorithme standard. Une angiographie coronarienne par TDM 256 coupes a été réalisée, avec reconstruction par algorithmes avec renforcement des bords et standard. L’épaisseur de paroi des tuteurs était mesurée par méthodes orthogonale (diamètres) et circonférentielle (circonférences). La qualité d’image des tuteurs était évaluée par échelle ordinale, et les données analysées par modèles linéaire mixte et régression logistique des cotes proportionnelles. L’épaisseur de paroi des tuteurs était inférieure avec l’algorithme avec renforcement des bords comparé à l’algorithme standard, avec les méthodes orthogonale (0,97±0,02 vs 1,09±0,03 mm, respectivement; p<0,001) et circonférentielle (1,13±0,02 vs 1,21±0,02 mm, respectivement; p<0,001). Le premier causait moins de surestimation par rapport à l’épaisseur nominale comparé au second, avec méthodes orthogonale (0,89±0,19 vs 1,00±0,26 mm, respectivement; p<0,001) et circonférentielle (1,06±0,26 vs 1,13±0,31 mm, respectivement; p=0,005) et diminuait de 6 % la surestimation. Les scores de qualité étaient meilleurs avec l’algorithme avec renforcement des bords (OR 3,71; IC 95% 2,33–5,92; p<0,001). En conclusion, la reconstruction des images avec l’algorithme avec renforcement des bords génère des parois de tuteurs plus minces, moins de surestimation, et de meilleurs scores de qualité d’image que l’algorithme standard.
Metallic artifacts can result in an artificial thickening of the coronary stent wall which can significantly impair computed tomography (CT) imaging in patients with coronary stents. The purpose of this study is to assess the in vivo visualization of coronary stent wall and lumen with an edge-enhancing CT reconstruction kernel, as compared to a standard kernel. This is a prospective cross-sectional study of 24 consecutive patients with 71 coronary stents, using a repeated measure design and blinded observers, approved by the Local Institutional Review Board. 256-slice CT angiography was used, as well as standard and edge-enhancing reconstruction kernels. Stent wall thickness was measured with orthogonal and circumference methods, averaging wall thickness from stent diameter and circumference measurements, respectively. Stent image quality was assessed on an ordinal scale. Statistical analysis used linear and proportional odds models. Stent wall thickness was inferior using the edge-enhancing kernel compared to the standard kernel, either with the orthogonal (0.97±0.02 versus 1.09±0.03 mm, respectively; p<0.001) or circumference method (1.13±0.02 versus 1.21±0.02 mm, respectively; p<0.001). The edge-enhancing kernel generated less overestimation from nominal thickness compared to the standard kernel, both with orthogonal (0.89±0.19 versus 1.00±0.26 mm, respectively; p<0.001) and circumference (1.06±0.26 versus 1.13±0.31 mm, respectively; p=0.005) methods. The average decrease in stent wall thickness overestimation with an edge-enhancing kernel was 6%. Image quality scores were higher with the edge-enhancing kernel (odds ratio 3.71, 95% CI 2.33–5.92; p<0.001). In conclusion, the edge-enhancing CT reconstruction kernel generated thinner stent walls, less overestimation from nominal thickness, and better image quality scores than the standard kernel.