Background: Heart damage from cancer therapy is a significant problem for survivors. Some of the most effective treatments, such as anthracyclines, cause heart toxicity that can lead to significant morbidity and mortality. Cardiotoxicity also contributes to the loss of promising cancer drugs in early development and is notoriously difficult to predict. This translational project employs parallel pre-clinical and clinical studies to explore circulating biomarkers and cardiac magnetic resonance imaging (CMR) during development of anthracycline associated cardiotoxicity with the aim of finding biomarkers to aid clinical decision making and enable forward/back translation. Methods: Pre-clinical work: A rat model of chronic anthracycline-induced cardiomyopathy was developed involving 8 weekly intravenous boluses of doxorubicin followed by a 4 week ‘washout’ period. A time course assessment of cardiac function using multiple MRI parameters was performed alongside a panel of circulating biomarkers measured prior to dosing. Clinical work: In parallel following ethical approval, 30 cancer patients receiving standard anthracycline chemotherapy were recruited. Serial CMR scans were performed using standard and new exploratory techniques before, during and after treatment and blood was taken to evaluate a similar panel of cardiotoxicity biomarkers using multiplex ELISA at corresponding time points. Results: Pre-clinical results: Systolic and diastolic function declined progressively, culminating in left ventricular dysfunction (LVEF < 50%) by 12 weeks. Myocardial electron microscopy revealed myofibrillar and mitochondrial damage after one dose and gross histopathological damage after 5 doses. Myocardial contrast enhancement and troponin I increased significantly after eight doses and preceded LV dysfunction. Extensive fibrosis was seen 1 month after drug cessation. Clinical results: LVEF declined progressively in all patients and 7 patients (23%) had persistent LV dysfunction 12 months after therapy. Troponin I elevations were seen towards the end of therapy and peak troponin I corresponded with LVEF decline. None of the other circulating biomarkers correlated strongly with outcome. Lower baseline extracellular volume (ECV) was associated with greater LVEF decline but little change in ECV was seen over time. Baseline dyssynchrony was associated with worse outcome and deteriorated with time alongside LVEF decline. Conclusions: Results suggest that troponin I and cardiac MRI are sensitive translational tools in drug induced cardiotoxicity. However, troponin I is a relatively late marker, peaking after substantial myocardial damage, too late to halt or change reatment. The imaging suggests that fibrosis and inflammation cannot be detected within a year of chemotherapy but baseline ECV and strain analysis may have a role in risk stratification.

No description available.

Improving imaging hardware, computational power, and algorithmic design are driving advances in interventional medical imaging. We lay the groundwork here for more effective use of machine learning and image registration in clinical electrophysiology. To achieve identification of atrial fibrosis using image data, we registered the electroanatomic map (EAM) data of atrial fibrillation (AF) patients undergoing pulmonary vein isolation (PVI) with MR (n = 16) or CT (n = 18) images. The relationship between image features and bipolar voltage was evaluated using single-parameter regression and random forest models. Random forest performed significantly better than regression, identifying fibrosis with area under the receiver operating characteristic curve (AUC) 0.746 (MR) and 0.977 (CT). This is the first evaluation of voltage prediction using image data. Next, we compared the character of native atrial fibrosis with ablation scar in MR images. Fourteen AF patients undergoing repeat PVI were recruited. EAM data from their first PVI was registered to the MR images acquired before the first PVI (‘pre-operative’) and before the second PVI ('post-operative' with respect to the first PVI). Non-ablation map points had similar characteristics in the two images, while ablation points exhibited higher intensity and more heterogeneity in post-operative images. Ablation scar is more strongly enhancing and more heterogeneous than native fibrosis. Finally, we addressed myocardial measurement in 3-D echocardiograms. The circular Hough transform was modified with a feature asymmetry filter, epicardial edges, and a search constraint. Manual and Hough measurements were compared in 5641 slices from 3-D images. The enhanced Hough algorithm was more accurate than the unmodified version (Dice coefficient 0.77 vs. 0.58). This method promises utility in segmentation-assisted cross-modality registration. By improving the information that can be extracted from medical images and the ease with which that information can be accessed, this progress will contribute to the advancing integration of imaging in electrophysiology.

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.

L'explosion récente de données d'imagerie cardiaque a été phénoménale. L'utilisation intelligente des grandes bases de données annotées pourrait constituer une aide précieuse au diagnostic et à la planification de thérapie. En plus des défis inhérents à la grande taille de ces banques de données, elles sont difficilement utilisables en l'état. Les données ne sont pas structurées, le contenu des images est variable et mal indexé, et les métadonnées ne sont pas standardisées. L'objectif de cette thèse est donc le traitement, l'analyse et l'interprétation automatique de ces bases de données afin de faciliter leur utilisation par les spécialistes de cardiologie. Dans ce but, la thèse explore les outils d'apprentissage automatique supervisé, ce qui aide à exploiter ces grandes quantités d'images cardiaques et trouver de meilleures représentations. Tout d'abord, la visualisation et l'interprétation d'images est améliorée en développant une méthode de reconnaissance automatique des plans d'acquisition couramment utilisés en imagerie cardiaque. La méthode se base sur l'apprentissage par forêts aléatoires et par réseaux de neurones à convolution, en utilisant des larges banques d'images, où des types de vues cardiaques sont préalablement établies. La thèse s'attache dans un deuxième temps au traitement automatique des images cardiaques, avec en perspective l'extraction d'indices cliniques pertinents. La segmentation des structures cardiaques est une étape clé de ce processus. A cet effet une méthode basée sur les forêts aléatoires qui exploite des attributs spatio-temporels originaux pour la segmentation automatique dans des images 3Det 3D+t est proposée. En troisième partie, l'apprentissage supervisé de sémantique cardiaque est enrichi grâce à une méthode de collecte en ligne d'annotations d'usagers. Enfin, la dernière partie utilise l'apprentissage automatique basé sur les forêts aléatoires pour cartographier des banques d'images cardiaques, tout en établissant les notions de distance et de voisinage d'images. Une application est proposée afin de retrouver dans une banque de données, les images les plus similaires à celle d'un nouveau patient. / The recent growth of data in cardiac databases has been phenomenal. Cleveruse of these databases could help find supporting evidence for better diagnosis and treatment planning. In addition to the challenges inherent to the large quantity of data, the databases are difficult to use in their current state. Data coming from multiple sources are often unstructured, the image content is variable and the metadata are not standardised. The objective of this thesis is therefore to simplify the use of large databases for cardiology specialists withautomated image processing, analysis and interpretation tools. The proposed tools are largely based on supervised machine learning techniques, i.e. algorithms which can learn from large quantities of cardiac images with groundtruth annotations and which automatically find the best representations. First, the inconsistent metadata are cleaned, interpretation and visualisation of images is improved by automatically recognising commonly used cardiac magnetic resonance imaging views from image content. The method is based on decision forests and convolutional neural networks trained on a large image dataset. Second, the thesis explores ways to use machine learning for extraction of relevant clinical measures (e.g. volumes and masses) from3D and 3D+t cardiac images. New spatio-temporal image features are designed andclassification forests are trained to learn how to automatically segment the main cardiac structures (left ventricle and left atrium) from voxel-wise label maps. Third, a web interface is designed to collect pairwise image comparisons and to learn how to describe the hearts with semantic attributes (e.g. dilation, kineticity). In the last part of the thesis, a forest-based machinelearning technique is used to map cardiac images to establish distances and neighborhoods between images. One application is retrieval of the most similar images.

No description available.

No description available.

L'imagerie par Résonance Magnétique (IRM) cardiaque est un domaine qui nécessite d'adapter la séquence au rythme du coeur, afin d'éviter le flou causé par un temps d'acquisition long devant les constantes de temps du mouvement. Ainsi, les temps séparant les impulsions radio-fréquence (RF) de la séquence sont aussi variables que les durées des cycles cardiaques sur lesquels on synchronise l'acquisition. Cela est cause d'imprécision sur l'image résultante, en particulier dans son caractère quantitatif. L'aimantation des spins n'est effectivement pas dans un état d'équilibre sur toute l'acquisition. La thèse présente deux axes principaux de recherche explorés ; le premier est une étude de l'impact de la variation du rythme cardiaque (présentée en outre dans le manuscrit) sur la mesure quantitative du temps de relaxation transversal T2. L'étude a été menée sur des objets fantômes et sur des volontaires sains. Deux méthodes de correction de la variation du rythme sont proposées, l'une basée sur la correction du signal au centre de l'espace de Fourier, l'autre basée sur une approche de reconstruction généralisée. Les résultats préliminaires sont encourageants, et des travaux ultérieurs seraient à entreprendre pour confirmer l'efficacité de ces méthodes. Ensuite, les variations temporelles du cycle cardiaque sont traitées à l'échelle d'un cycle, et nous proposons une méthode de mise en coïncidence des différents segments de l'électrocardiogramme (ECG) basée sur la déformation de l'ECG dans l'IRM probablement par effet magnétohydrodynamique. Cette méthode est mise au service de l'imagerie dans le cadre d'une séquence cinétique CINE dans laquelle une meilleure mise en correspondance des segments de cycles cardiaques successifs devrait permettre de gagner en qualité d'image, à condition d'avoir des résolutions spatiale et temporelle suffisamment fines. Les résultats apportés au cours de cette thèse sont préliminaires à de futures recherches nécessaires dans le domaine temporel de la séquence, beaucoup moins traité que le mouvement des organes / Cardiac Magnetic Resonance Imaging (MRI) requires to adapt the sequence to heart rate, so as to avoid the blur caused by the acquisition time longer than motion time constants. Thus, times between sequence radiofrequency pulses are as much variable as synchronization cardiac cycles durations. It causes imprecision on the resulting image, particularly for quantification. In fact, spins magnetization is not in a steady state during the acquisition. Two main research axis are presented in this thesis; the first one is a study of the impact of heart rate variation (described in the manuscript) on the transversal relaxation time T2 quantitative measurement. The study was lead on both phantom objects and on healthy volunteers. Two correction methods for heart rate variation are proposed, one based on the correction of the signal of the central line of the k-space, the other one based on a generalized reconstruction approach. First results are encouraging, and further works should be lead to confirm the methods efficacity. Then, heart rate variations are treated inside the cardiac cycle, and we propose a method to match the different segments of the electrocardiogram (ECG), based on the ECG deformation in the MR scanner probably due to by magnetohydrodynamic effect. This method is applied on imaging with a CINE kinetic sequence in which a better successive cardiac cycles segments matching should enable to improve image quality, at the condition to have sharp enough spatial and temporal resolutions. Results brought in this thesis are preliminary to necessary future researches in the sequence time domain, largely less addressed than organ motion

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.

Motion artefacts due to cardiac and respiratory motion present a daily challenge in cardiac Magnetic Resonance Imaging (MRI), and many different motion correction procedures are used in clinical routine imaging. To reduce motion artefacts further, patients are required to hold their breath during parts of the data acquisition, which is physically straining – especially when done repetitively. Self-Gating (SG) is a method that extracts cardiac and respiratory motion information from the MRI data in the form of signals, called SG signals, and uses them to divide the data into the specific cardiac and respiratory phases it was acquired from. This method both avoids motion artefacts and allow for free-breathing acquisition. This project’s goal was to find a method for extracting cardiac and respiratory SG signals from MRI data. The data was acquired with a golden angle radial acquisition method for 3-dimensional (3D) scans. Extraction of the raw signal was tested for both raw k-space data and high temporal resolution image series, where the images were reconstructed using a sliding window reconstruction. Filters were then applied to isolate the cardiac and respiratory information, to create separate cardiac and respiratory SG signals. Thereafter trigger points marking the beginning of the cardiac and respiratory cycles were generated. The trigger points were compared against ECG and respiratory trigger points provided by the MR scanner. The conclusion was that the SG signals based on k-space data was functional on the scans from the evaluated subjects and the most effective choice of the two options, but image based SG signals may prove to be functional after further studies. / Rörelseartefakter på grund av hjärt- och respirationsrörelser är idag vardagliga utmaningar inom magnetresonanstomografi (MR) av hjärtat, och många olika metoder används för att eliminera rörelseartefakterna. Patienterna behöver dessutom hålla andan under delar av dataupptagningen, vilket är fysiskt ansträngande – speciellt när det sker upprepade gånger. Self-Gating (SG) är en metod som extraherar information hjärt- och respirationsrytm från MR-datan i form av signaler, kallade SG signaler, och använder dem för att dela in datan i de specifika hjärt- respektive respirationsfaser som var när datan upptogs. Denna metod både undviker rörelseartefakter och tillåter fri andning under dataupptagningen. Målet med det här projektet var att hitta en metod för att extrahera SG signaler för hjärt- och respirationsrytm från MR-data. Datan samlades in med en golden angle radial-upptagning för 3- dimensionella (3D) scanningar. Extraheringen av den råa signalen testades på både rå k-space data och på bildserier av 3D-bilder med hög tidsupplösning, där bilderna var rekonstruerade med en sliding window rekonstruktion. Därefter applicerades filter för att isolera hjärt- och respirationsinformationen, för att få separata SG signaler med endast hjärt- respektive respirationsrytmer. Till slut genererades triggerpunkter för att markera början av hjärt- respektive respirationscyklerna. Dessa jämfördes med triggerpunkter uppmätta med EKG och andningskudde i magnetkameran. Slutsatsen för projektet var att SG signalerna som baserades på k-space data var funktionell för de scanningar som testades och det mest effektiva alternativet, men SG signalerna som baserades på bilder kan visa sig fungera efter mer studier.