In recent years, researchers have increasingly used magnetic resonance imaging (MRI) to study temporal skeletal muscle changes using gradient echo (GRE) echo planar imaging (EPI). These studies, typically involving exercise or ischemic challenges, have differentiated healthy subjects from athletic or unhealthy populations, such as those with peripheral vascular disease. However, the analysis methodologies have been lacking. In this thesis, two sessions of post-exercise GRE EPI data were collected from six subjects' lower legs using a 3 Tesla MRI scanner and a custom built ergometer. Past studies used common medical imaging software for motion correction. This work shows that such tools degrade leg image data by introducing motion, increasing root mean squared error in rest data by 22%. A new approach decreased it by 12%. EPI distortion correction in muscle images was also achieved, with the correlation ratio of functional and structural images increasing by up to 8%. In addition, a brief but intense artifact in GRE EPI muscle images results from muscle tissue moving in and out of the imaged volume. This through-plane artifact was successfully modelled as a mono-exponential decay for regression analysis, increasing the utility of the residual signal. The regression parameters were also leveraged to produce muscle displacement maps, identifying 44% of voxels as displaced. The maps were validated in a motion phantom and in-vivo using ultrasound. Finally, independent component analysis (ICA) was applied to post-exercise GRE EPI images to detect features in a data-driven, multivariate way and improve on conventional ROI selection methods. ICA produced parametric maps that were spatially correlated to working muscles from every trial (most with |R| > 0.4). The components were also separated from the susceptibility, motion, and blood vessel signals, and temporally reliable within individuals. These methodological advances represent increased rigour in the analysis of muscle GRE EPI images. / Thesis / Doctor of Philosophy (PhD) / Adequate blood circulation to muscles is important for good health. Researchers have used magnetic resonance imaging (MRI) techniques to assess blood and oxygen supply to muscles. The work in this thesis improves upon the analysis methods in prior work, especially in the areas of motion correction of the images and selection of individual muscle regions for analysis. Previous techniques could sometimes make motion in muscle images worse. This work provides valuable motion and distortion correction for muscle imaging, ensuring that measurements truly reflect muscle physiology. It also describes a method to remove an unwanted signal from post-exercise muscle data, and create a map of the internal muscle motion that occurred. Finally, an advanced mathematical technique was used to extract signals of interest and important spatial features from muscle image data automatically. The technique produced reliable results within and among subjects.

Les propriétés mécaniques des tissus biologiques sont des paramètres importants en médecine : ce sont des biomarqueurs du fonctionnement normal ou pathologique d'un tissu. En effet, ces propriétés peuvent être affectées par certaines conditions mécaniques telles que l'application d'une contrainte externe, comme l'hypertension ou un traumatisme, mais également par la présence de certaines maladies, telles que le cancer, la fibrose, l’inflammation, la maladie d'Alzheimer, ou bien tout simplement avec l'âge. La palpation réalisée par le médecin permet de discerner ces changements mais ce geste est qualitatif et ne peut accéder à des organes profonds. L'élastographie-IRM reste une méthode quantitative, robuste, d'une grande précision, qui permet de sonder l'élasticité et la viscosité des tissus. Elle consiste à mesurer le champ de déplacement d'une onde de cisaillement induite dans l'organe ciblé par une technique IRM en contraste de phase. Les modules viscoélastiques sont alors déduits après inversion de l'équation d'onde. Malgré cela, la justesse de cette technique n'a pas encore été pleinement établie. L'élastographie-IRM est en cours d'implémentation en routine clinique sur des patients atteints de maladies hépatiques chroniques ou bien pour caractériser des tumeurs dans le cas de cancer du sein. L'application aux autres organes protégés, tels que le cerveau ou les poumons, reste encore du domaine de la recherche à cause de la difficulté d'y induire des ondes mécaniques (protection naturelle de la boîte crânienne ou de la cage thoracique). C'est dans ce contexte qu'intervient un volet de mon travail de thèse : la mise en place, la caractérisation et l'optimisation d'un système induisant des ondes mécaniques dans les organes profonds. L’approche originale suivie a été d’utiliser les voies naturelles permettant d’amener l’onde de pression aux poumons ou bien à l’encéphale, différente des approches classiques consistant à traverser les barrières protectrices. Ce générateur d'onde de pression nous a permis d'obtenir des amplitudes d'onde allant de 6 µm à 30 µm dans l'ensemble du cerveau, amplitudes suffisantes afin d'en déduire les modules viscoélastiques du cerveau entier. D'autre part, un travail important s'est attaché à la réalisation d'un schéma original de correction des mouvements du patient en élastographie-IRM. Nous avons mis en évidence comment ces mouvements peuvent entraîner une discordance des composantes du champ de déplacement, nécessitant alors d'être corrigées. La correction proposée est composée d'une première étape dont la finalité est de recaler spatialement l'ensemble des volumes acquis, puis d'une seconde étape permettant de rétablir les composantes du champ de déplacement dans la même base orthonormée. Nous avons évalué numériquement et expérimentalement le biais induit quand aucunes corrections n'étaient appliquées sur ces données ainsi que l'apport de ces deux étapes de correction. Un travail préliminaire sur l'étude de la reproductibilité des acquisitions (phase en particulier) a été nécessaire. Enfin, l'ensemble des résultats de ces deux volets nous ont permis de réaliser des acquisitions d'élastographie du cerveau complet et d'obtenir des cartes du champ de déplacement de qualité. Ainsi, nous avons pu montrer la tendance des ondes mécaniques à suivre les directions privilégiées des fibres du cerveau, résultats que nous avons commencé à confronter aux observations faites en DTI. / Mechanical properties of biological tissues are important parameters in medicine: they are normal or pathological function biomarkers of tissue. Indeed, these properties can be affected by some mechanical conditions such as the application of an external constraint, like hypertension or trauma, but also by the presence of certain diseases, such as cancer, fibrosis, inflammation, Alzheimer’s disease, or simply with age. Palpation performed by the physician can detect these changes but this gesture is qualitative and can not access deep organs. MR-elastography remains a quantitative and robust method of high precision, which probes elasticity and viscosity of tissues. It consists in measuring the displacement field of a shear wave induced in the target organ by a phase contrast based MRI technique. The viscoelastic moduli are deducted after inversion of the wave equation. Nevertheless, the accuracy of this technique has not yet been fully established. MR-elastography is being implemented in routine clinical practice for patients with chronic liver diseases or to characterize tumors in the case of breast cancer. Application to other protected organs, such as the brain or lungs, is still in research area because of the difficulty to induce mechanical waves (natural protection of the skull or the rib cage). It is in this context that a part of my thesis work is involved: the establishment, characterization and optimization of a system inducing mechanical waves in deep organs. The original approach was to use anatomical pathways for bringing the pressure waves to the lungs or the brain, different from conventional approaches of traversing the protective barriers. This pressure wave generator allowed us to obtain wave amplitudes ranging from 6 µm to 30 µm in the whole brain, sufficient amplitudes to deduce the whole brain viscoelastic moduli. On the other hand, an important work has focused on the realization of an original scheme of patient motions correction in MR-elastography. We have brought out how these motions can cause a mismatch of the displacement field components, which need to be corrected. The proposed correction is composed of a first step whose purpose is to spatially realign all acquired volumes, then a second step to restore the displacement field components in the same orthonormal basis. We numerically and experimentally evaluated the bias when no corrections were applied to these data and the contribution of these two correction steps. A preliminary work on the study of the acquisitions reproducibility (particularly phase) was necessary. Finally, all the results of these two components have allowed us to realize elastography acquisitions of the whole brain and obtain quality displacement field maps. Thus, we showed the trend of mechanical waves to follow the brain fibers preferred directions, results that we started to compare to the observations made by DTI.

Positronemissionstomografi är en väl använd modalitet som kan hjälpa till att ge information om olika organs tillstånd. Ett problem som kan uppstå när man utför undersökningar är att patientrörelser, i synnerhet andning, förvränger de resulterande bilderna. Detta är ett vanligt problem och kan leda till komplikationer som inkorrekt diagnos och i sin tur felbehandling. Om rörelsen är för kraftig kan det till och med göra bilden värdelös, vilket tvingar patienten att genomgå en ny examination som är kostsam både för patientens hälsa och för sjukhuset som utför undersökningen. För att förhindra att detta inträffar har flera studier gjorts för att anpassa bilderna så att även om patienten rör sig skulle undersökningen fortfarande ge acceptabla bilder. Detta projekt är en fortsättning på det arbete som gjorts av en annan student på CBH-skolan vid KTH, som för sin masteruppsats tränade en djupinlärningsmodell att korrigera för patientrörelser. Denna modell tränades uteslutande på PET-fantomer och inte attenuerad data som genererats i simuleringar. I detta projekt prövades och implementerades tre olika metoder för att ta fram attenueringsmappar med hjälp av modellen för att utföra både rörelse- och attenueringskorrektion. Resultaten visade att metoden som direkt tillämpade deformationen mellan attenueringskarta och respektive PET-gate var överlägsen de andra, men den kunde fortfarande inte återge korrigerade bilder där lesionerna som fanns på fantomen är tydligt synliga. / Positron emission tomography is a widely used modality that can help provide information about how various parts of the body are functioning. An issue that can occur when performing these examinations is patient movement, usually breathing, distorting the resulting images. This is a common issue and can lead to complications such as misdiagnosis and in turn mistreatment. If the movement is too significant it can even render the scan useless, forcing the patient to undergo a new scan which is costly both for the patient’s health and for the hospital performing the examination. To prevent this from happening several methods have been tried to adapt the images so that even if the patient moved the scan would still produce acceptable images.  This project is a continuation of the work done by another student on the CBH-school at KTH who, for his master’s thesis, trained a deep-learning model to correct for patient movement. This model was trained exclusively on PET phantoms and not attenuated data generated in simulations. In this project three different methods were tested and implemented to acquire attenuation maps that could be used to perform movement- and attenuation correction. The results showed that the method that directly applied the deformation between attenuation map and respective PET-gate was superior to the others, but it could still not reconstruct corrected images with the lesions that were originally visible on the phantom.

Positron Emission Tomography (PET) is pivotal in medical imaging but is prone to artifactsfrom physiological movements, notably respiration. These motion artifacts both degradeimage quality and compromise precise attenuation correction. To counteract this, gatingstrategies partition PET data in synchronization with respiratory cycles, ensuring each gatenearly represents a static phase. Additionally, a 3D deep learning image registration modelcan be used for inter-gate motion correction, maximizing the use of the full acquired data. Thisstudy aimed to implement and evaluate two gating strategies: an external device-based approachand a data-driven centroid-of-distribution (COD) trace algorithm, and assess their impact on theperformance of the registration model. Analysis of clinical data from four subjects indicated thatthe external device approach outperformed its data-driven counterpart, which faced challengesin real-patient settings. Post motion compensation, both methods achieved results comparableto state-of-the-art reconstructions, suggesting the deep learning model addressed some data-driven method limitations. However, the motion corrected outputs did not exhibit significantimprovements in image quality over state-of-the-art standards. / Positronemissionstomografi (PET) är fundamentalt inom medicinsk avbildning men påverkasav artefakter orsakade av fysiologiska rörelser, framför allt andning. Dessa artefakter påverkarbildkvaliteten negativt och försvårar korrekt attenueringskorrigering. För att motverka dettakan tekniker för rörelsekorrigering tillämpas. Dessa innefattar gating-tekniker där PET-dataförst synkroniseras med andningscykeln för att därefter segmenterateras i olika så kalladegater som representerar en specifick respiratorisk fas. Vidare kan en 3D djupinlärningsmodellanvändas för att korrigera för rörelserna mellan gaterna, vilket optimerar användningen av allinsamlad data. Denna studie implementerade och undersökte två gating-tekniker: en externenhetsbaserad metod och en datadriven ”centroid-of-distribution (COD)” spår-algoritm, samtanalyserade hur dessa tekniker påverkar prestandan av bildregistreringsmodellen. Utifrånanalysen av kliniska data från fyra patienter visade sig metoden med den externa enhetenvara överlägsen den datadrivna metoden, som hade svårigheter i verkliga patient-situationer.Trots detta visade bildregistreringsmodellen potential att delvis kompensera för den datadrivnametodens begränsningar, då resultatet från båda strategeierna var jämförbara med befintligaklinisk bildrekonstruktion. Dock kunde ingen markant förbättring i bildkvalitet urskiljas av derörelsekorrigerade bilderna jämfört med nuvarande toppstandard.

No description available.

L'Imagerie par Résonance Magnétique (IRM) présente deux caractéristiques principales. La première, sa capacité à manipuler le contraste, constitue son principal avantage par rapport aux autres modalités d'imagerie. Cela permet d'obtenir des informations complémentaires pour une meilleure détectabilité et une meilleure précision dans le diagnostic. Cela est particulièrement appréciable pour les pathologies du myocarde. La seconde caractéristique de l'IRM est également l'un de ces principaux inconvénients : le processus d'acquisition est relativement lent. De ce fait, les mouvements du patient constituent un obstacle important puisqu'ils perturbent ce processus d'acquisition, ce qui se traduit par des artéfacts dans l'image reconstruite. L'imagerie cardiaque et abdominale sont donc particulièrement sensibles à cette problématique du mouvement. L'objectif de cette thèse est donc de proposer une méthode de correction de mouvement intégrable dans un contexte multi-contraste. Nous avons étudié dans un premier temps la question de la correction de mouvement seule. Pour cela, nous nous sommes plus particulièrement intéressés à la méthode GRICS déjà développée au laboratoire IADI. Cette méthode permet la reconstruction conjointe d'une image sans artéfact et d'un modèle de mouvement non rigide permettant de corriger les déplacements qui surviennent pendant l'acquisition. Le premier apport majeur de cette thèse a consisté à améliorer la méthode GRICS, notamment pour l'adapter à l'imagerie volumique 3D. Il s'agit d'une nouvelle méthode de régularisation adaptative particulièrement adaptée au problème inverse posé dans GRICS. Le second apport majeur de cette thèse a consisté à gérer la correction de mouvement GRICS de manière conjointe sur des acquisitions présentant des contrastes différents. Il s'agit de concevoir l'examen IRM comme un tout et d'exploiter au mieux les informations partagées entre les différents contrastes. Toutes ces méthodes ont été appliquées et validées par des simulations, des tests sur fantôme, sur volontaires sains et sur des patients dans la cadre d'études cliniques. L'application cardiaque a été particulièrement visée. Les méthodes développées ont permis d'améliorer le processus d'acquisition et de reconstruction dans le contexte clinique réel / Magnetic Resonance Imaging (MRI) has two main features. The first one, its ability to manipulate contrast, is a major advantage compared to the other imaging modalities. It allows to access complementary information for a better detectability and a diagnostic more accurate. This is especially useful for myocardium pathologies. The second feature of MRI is also one of its main drawbacks: the acquisition process is slow. Therefore, patient motion is a significant obstacle because it disturbs the acquisition process, which leads to artifacts in the reconstructed image. Cardiac and thoracic imaging are particularly sensitive to this motion issue. The aim of this thesis is to develop a new motion correction method that can be integrated in a multi-contrast workflow. In a first phase, we studied apart the motion correction problem. To do so, we focused more particularly on the GRICS method which was already developed in the IADI laboratory. This method allows the joint reconstruction of an image free from artifact and a non-rigid motion model that describes the displacements occurring during the acquisition. The first major contribution of this thesis is an improvement of the GRICS method consisting mainly in adapting it to the 3D imaging. This was achieved with a new adaptive regularization method that perfectly suits the inverse problem posed in GRICS. The second major contribution of this thesis consists in the simultaneous management of the motion correction on multiple acquisitions with different contrasts. To do so, the MRI examination is considered as a whole. Thus we make the most of information shared between the different contrasts. All these methods have been applied and validated by simulations, tests on phantom, on healthy volunteers and on patients as part of clinical studies. We aimed more particularly at cardiac MR. Finally the developed methods improve the acquisition and reconstruction workflow in the framework of a real clinical routine

As a modern tomographic technique, Positron-Emission-Tomography (PET) enables non-invasive imaging of metabolic processes in living organisms. It allows the visualization of malfunctions which are characteristic for neurological, cardiological, and oncological diseases. Chemical tracers labeled with radioactive positron emitting isotopes are injected into the patient and the decay of the isotopes is then observed with the detectors of the tomograph. This information is used to compute the spatial distribution of the labeled tracers. Since the spatial resolution of PET devices increases steadily, the whole sensitive process of tomograph imaging requires minimizing not only the disturbing effects, which are specific for the PET measurement method, such as random or scattered coincidences, but also external effects like body movement of the patient. Methods to correct the influences of such patient movement have been developed in previous studies at the PET center, Rossendorf. These methods are based on the spatial correction of each registered coincidence. However, the large amount of data and the complexity of the correction algorithms limited the application to selected studies. The aim of this thesis is to optimize the correction algorithms in a way that allows movement correction in routinely performed PET examinations. The object-oriented development in C++ with support of the platform independent Qt framework enables the employment of multiprocessor systems. In addition, a graphical user interface allows the use of the application by the medical assistant technicians of the PET center. Furthermore, the application provides methods to acquire and administrate movement information directly from the motion tracking system via network communication. Due to the parallelization the performance of the new implementation demonstrates a significant improvement. The parallel optimizations and the implementation of an intuitive usable graphical interface finally enables the PET center Rossendorf to use movement correction in routine patient investigations, thus providing patients an improved tomograph imaging. / Die Positronen-Emissions-Tomographie (PET) ist ein modernes medizinisches Diagnoseverfahren, das nichtinvasive Einblicke in den Stoffwechsel lebender Organismen ermöglicht. Es erfasst Funktionsstörungen, die für neurologische, kardiologische und onkologische Erkrankungen charakteristisch sind. Hierzu werden dem Patienten radioaktive, positronen emittierende Tracer injiziert. Der radioaktive Zerfall der Isotope wird dabei von den umgebenden Detektoren gemessen und die Aktivitätsverteilung durch Rekonstruktionsverfahren bildlich darstellbar gemacht. Da sich die Auflösung solcher Tomographen stetig verbessert und somit sich der Einfluss von qualitätsmindernden Faktoren wie z.B. das Auftreten von zufälligen oder gestreuten Koinzidenzen erhöht, gewinnt die Korrektur dieser Einflüsse immer mehr an Bedeutung. Hierzu zählt unter anderem auch die Korrektur der Einflüsse eventueller Patientenbewegungen während der tomographischen Untersuchung. In vorangegangenen Studien wurde daher am PET Zentrum Rossendorf ein Verfahren entwickelt, um die nachträgliche listmode-basierte Korrektur dieser Bewegungen durch computergestützte Verfahren zu ermöglichen. Bisher schränkte der hohe Rechenaufwand den Einsatz dieser Methoden jedoch ein. Diese Arbeit befasst sich daher mit der Aufgabe, durch geeignete Parallelisierung der Korrekturalgorithmen eine Optimierung dieses Verfahrens in dem Maße zu ermöglichen, der einen routinemässigen Einsatz während PET Untersuchungen erlaubt. Hierbei lässt die durchgeführte objektorientierte Softwareentwicklung in C++ , unter Zuhilfenahme des plattformübergreifenden Qt Frameworks, eine Nutzung von Mehrprozessorsystemen zu. Zusätzlich ermöglicht eine graphische Oberfläche die Bedienung einer solchen Bewegungskorrektur durch die medizinisch technischen Assistenten des PET Zentrums. Um darüber hinaus die Administration und Datenakquisition der Bewegungsdaten zu ermöglichen, stellt die entwickelte Anwendung Funktionen bereit, die die direkte Kommunikation mit dem Bewegungstrackingsystem erlauben. Es zeigte sich, dass durch die Parallelisierung die Geschwindigkeit wesentlich gesteigert wurde. Die parallelen Optimierungen und die Implementation einer intuitiv nutzbaren graphischen Oberfläche erlaubt es dem PET Zentrum nunmehr Bewegungskorrekturen innerhalb von Routineuntersuchungen durchzuführen, um somit den Patienten ein verbessertes Bildgebungsverfahren bereitzustellen.

L'imagerie TEP permet de mesurer et visualiser les changements de la distribution biologique des radiopharmaceutiques au sein des organes d'intérêt au court du temps. Ce suivi temporel offre des informations très utiles concernant les processus métaboliques et physiologiques sous-jacents, qui peuvent être extraites grâce à différentes techniques de modélisation cinétique. De plus, un autre avantage de la prise en compte de l'information temporelle dans les acquisitions TEP pour les examens en oncologie thoracique concerne le suivi des mouvements respiratoires. Ces acquisitions permettent de mettre en place des protocoles et des méthodologies visant à corriger leurs effets néfastes à la quantification, et les artefacts associés. L'objectif de ce projet est de développer une méthode de reconstruction permettant de combiner et mettre en oeuvre d'une part les corrections nécessaires à la quantification des données en TEP, et d'autre part la modélisation de la biodistribution du radiotraceur au cours du temps permettant d'obtenir des images paramétriques pour l'oncologie thoracique. Dans un premier temps, une méthodologie de correction des effets de volume partiel intégrant, dans le processus de reconstruction, une déconvolution de Lucy-Richardson associée à un débruitage dans le domaine des ondelettes, a été proposée. Une seconde étude a été consacrée au développement d'une méthodologie combinant une régularisation temporelle des données par l'intermédiaire d'un ensemble de fonctions de base temporelles, avec une méthode de correction des mouvements respiratoires basée sur un modèle élastique. Enfin, dans une troisième étape, le modèle cinétique de Patlak a été intégré dans un algorithme de reconstruction dynamique, et associé à la correction de mouvement afin de permettre la reconstruction directe d'images paramétriques de données thoraciques soumises au mouvement respiratoire. Les paramètres de transformation élastique pour la correction de mouvement ont été calculés à partir des images TEP d'intervalles synchronisés par rapport à l'amplitude de la respiration du patient. Des simulations Monte-Carlo d'un fantôme 4D géométrique avec plusieurs niveaux de statistiques, et du fantôme anthropomorphique NCAT intégrant des courbes d'activités temporelles réalistes pour les différents tissus, ont été réalisées afin de comparer les performances de la méthode de reconstruction paramétrique développée dans ce travail avec une approche 3D standard d'analyse cinétique. L'algorithme proposé a ensuite été testé sur des données cliniques de patients présentant un cancer bronchique non à petites cellules. Enfin, après la validation indépendante de l'algorithme de correction des effets de volume partiel d'une part, et de la reconstruction 4D incorporant la régularisation temporelle d'autre part, sur données simulées et cliniques, ces deux méthodologies ont été associées afin d'optimiser l'estimation de la fonction d'entrée à partir d'une région sanguine des images reconstruites. Les résultats de ce travail démontrent que l'approche de reconstruction paramétrique proposée permet de conserver un niveau de bruit stable dans les régions tumorales lorsque la statistique d'acquisition diminue, contrairement à l'approche d'estimation 3D pour laquelle le niveau de bruit constaté augmente. Ce résultat est intéressant dans l'optique d'une réduction de la durée des intervalles de la reconstruction 4D, permettant ainsi de réduire la durée totale de l'acquisition 4D. De plus, l'utilisation des fonctions d'entrée estimées avec les méthodes de régularisation temporelle proposées ont conduit à améliorer l'estimation des paramètres de Patlak. Enfin, la correction élastique du mouvement amène à une diminution du biais d'estimation des deux paramètres de Patlak, en particulier sur les tumeurs de petites dimensions situées dans des régions sensibles au mouvement respiratoire. / Positron emission tomography (PET) is now considered as the gold standard and the main tool for the diagnosis and therapeutic monitoring of oncology patients, especially due to its quantitative aspects. With the advent of multimodal imaging in combined PET and X-ray CT systems, many methodological developments have been proposed in both pre-processing and data acquisition, image reconstruction, as well as post-processing in order to improve the quantification in PET imaging. Another important aspect of PET imaging is its high temporal resolution and ability to perform dynamic acquisitions, benefiting from the high sensitivity achieved with current systems. PET imaging allows measuring and visualizing changes in the biological distribution of radiopharmaceuticals within the organ of interest over time. This time tracking provides valuable information to physicians on underlying metabolic and physiological processes, which can be extracted using pharmacokinetic modeling. The objective of this project is, by taking advantage of dynamic data in PET/CT imaging, to develop a reconstruction method combining in a single process all the correction methodology required to accurately quantify PET data and, at the same time, include a pharmacokinetic model within the reconstruction in order to create parametric images for applications in oncology. In a first step, a partial volume effect correction methodology integrating, within the reconstruction process, the Lucy-Richardson deconvolution algorithm associated with a wavelet-based denoising method has been introduced. A second study focused on the development of a 4D reconstruction methodology performing temporal regularization of the dataset through a set of temporal basis functions, associated with a respiratory motion correction method based on an elastic deformation model. Finally, in a third step, the Patlak kinetic model has been integrated in a dynamic image reconstruction algorithm and associated with the respiratory motion correction methodology in order to allow the direct reconstruction of parametric images from dynamic thoracic datasets affected by the respiratory motion. The elastic transformation parameters derived for the motion correction have been estimated from respiratory-gated PET images according to the amplitude of the patient respiratory cycle. Monte-carlo simulations of two phantoms, a 4D geometrical phantom, and the anthropomorphic NCAT phantom integrating realistic time activity curves for the different tissues, have been performed in order to compare the performances of the proposed 4D parametric reconstruction algorithm with a standard 3D kinetic analysis approach. The proposed algorithm has then been assessed on clinical datasets of several patients with non small cell lung carcinoma. Finally, following the prior validation of the partial volume effect correction algorithm on one hand, and the 4D reconstruction incorporating the temporal regularization on the other hand, on simulated and clinical datasets, these two methodologies have been associated within the 4D reconstruction algorithm in order to optimize the estimation of image derived input functions. The results of this work show that the proposed direct parametric approach allows to maintain a similar noise level in the tumor regions when the statistic decreases, contrary to the 3D estimation approach for which the observed noise level increases. This result suggests interesting perspectives for the reduction of frame duration reduction of 4D reconstruction, allowing a reduction of the total 4D acquisition duration. In addition, the use of input function estimated with the developed temporal regularization methods led to the improvement of the Patlak parameters estimation. Finally, the elastic respiratory motion correction led to a diminution of the estimation bias of both Patlak parameters, in particular for small lesions located in regions affected by the respiratory motion.

Positron emission tomography (PET) is a well-established functional imaging method used in nuclear medicine. It allows for retrieving information about biochemical and physiological processes in vivo. The currently possible spatial resolution of PET is about 5 mm for brain acquisitions and about 8 mm for whole-body acquisitions, while recent improvements in image reconstruction point to a resolution of 2 mm in the near future. Typical acquisition times range from minutes to hours due to the low signal-to-noise ratio of the measuring principle, as well as due to the monitoring of the metabolism of the patient over a certain time. Therefore, patient motion increasingly limits the possible spatial resolution of PET. In addition, patient immobilisations are only of limited benefit in this context. Thus, patient motion leads to a relevant resolution degradation and incorrect quantification of metabolic parameters. The present work describes the utilisation of a novel motion compensation method for clinical brain PET acquisitions. By using an external motion tracking system, information about the head motion of a patient is continuously acquired during a PET acquisition. Based on the motion information, a newly developed event-based motion compensation algorithm performs spatial transformations of all registered coincidence events, thus utilising the raw data of a PET system - the so-called `list-mode´ data. For routine acquisition of this raw data, methods have been developed which allow for the first time to acquire list-mode data from an ECAT Exact HR+ PET scanner within an acceptable time frame. Furthermore, methods for acquiring the patient motion in clinical routine and methods for an automatic analysis of the registered motion have been developed. For the clinical integration of the aforementioned motion compensation approach, the development of additional methods (e.g. graphical user interfaces) was also part of this work. After development, optimisation and integration of the event-based motion compensation in clinical use, analyses with example data sets have been performed. Noticeable changes could be demonstrated by analysis of the qualitative and quantitative effects after the motion compensation. From a qualitative point of view, image artefacts have been eliminated, while quantitatively, the results of a tracer kinetics analysis of a FDOPA acquisition showed relevant changes in the R0k3 rates of an irreversible reference tissue two compartment model. Thus, it could be shown that an integration of a motion compensation method which is based on the utilisation of the raw data of a PET scanner, as well as the use of an external motion tracking system, is not only reasonable and possible for clinical use, but also shows relevant qualitative and quantitative improvement in PET imaging. / Die Positronen-Emissions-Tomographie (PET) ist ein in der Nuklearmedizin etabliertes funktionelles Schnittbildverfahren, das es erlaubt Informationen über biochemische und physiologische Prozesse in vivo zu erhalten. Die derzeit erreichbare räumliche Auflösung des Verfahrens beträgt etwa 5 mm für Hirnaufnahmen und etwa 8 mm für Ganzkörperaufnahmen, wobei erste verbesserte Bildrekonstruktionsverfahren eine Machbarkeit von 2 mm Auflösung in Zukunft möglich erscheinen lassen. Durch das geringe Signal/Rausch-Verhältnis des Messverfahrens, aber auch durch die Tatsache, dass der Stoffwechsel des Patienten über einen längeren Zeitraum betrachtet wird, betragen typische PET-Aufnahmezeiten mehrere Minuten bis Stunden. Dies hat zur Folge, dass Patientenbewegungen zunehmend die erreichbare räumliche Auflösung dieses Schnittbildverfahrens limitieren. Eine Immobilisierung des Patienten zur Reduzierung dieser Effekte ist hierbei nur bedingt hilfreich. Es kommt daher zu einer relevanten Auflösungsverschlechterung sowie zu einer Verfälschung der quantifizierten Stoffwechselparameter. Die vorliegende Arbeit beschreibt die Nutzbarmachung eines neuartigen Bewegungskorrekturverfahrens für klinische PET-Hirnaufnahmen. Mittels eines externen Bewegungsverfolgungssystems wird während einer PET-Untersuchung kontinuierlich die Kopfbewegung des Patienten registriert. Anhand dieser Bewegungsdaten führt ein neu entwickelter event-basierter Bewegungskorrekturalgorithmus eine räumliche Korrektur aller registrierten Koinzidenzereignisse aus und nutzt somit die als "List-Mode" bekannten Rohdaten eines PET Systems. Für die Akquisition dieser Daten wurden eigens Methoden entwickelt, die es erstmals erlauben, diese Rohdaten von einem ECAT Exact HR+ PET Scanner innerhalb eines akzeptablen Zeitraumes zu erhalten. Des Weiteren wurden Methoden für die klinische Akquisition der Bewegungsdaten sowie für die automatische Auswertung dieser Daten entwickelt. Ebenfalls Teil der Arbeit waren die Entwicklung von Methoden zur Integration in die klinische Routine (z.B. graphische Nutzeroberflächen). Nach der Entwicklung, Optimierung und Integration der event-basierten Bewegungskorrektur für die klinische Nutzung wurden Analysen anhand von Beispieldatensätzen vorgenommen. Es zeigten sich bei der Auswertung sowohl der qualitativen als auch der quantitativen Effekte deutliche Änderungen. In qualitativer Hinsicht wurden Bildartefakte eliminiert; bei der quantitativen Auswertung einer FDOPA Messung zeigte sich eine revelante Änderung der R0k3 Einstromraten eines irreversiblen Zweikompartment-Modells mit Referenzgewebe. Es konnte somit gezeigt werden, dass eine Integration einer Bewegungskorrektur unter Zuhilfenahme der Rohdaten eines PET Systems sowie unter Nutzung eines externen Verfolgungssystems nicht nur sinnvoll und in der klinischen Routine machbar ist, sondern auch zu maßgeblichen qualitativen und quantitativen Verbesserungen in der PET-Bildgebung beitragen kann.

Exchange of mass (water vapor), momentum, and energy between atmosphere andocean has profound influence on weather and climate. This exchange takes place at the air-sea interface, which is part of the marine atmospheric boundary layer. Various empirical relations are being used for estimating these fluxes in numericalweather and climate models but their accuracies are not sufficiently verified or tested over the Indian Ocean. The main difficulty is that vast areas of open oceans are not easily accessible. The marine environment is very corrosive and unattended long term and accurate measurements are extremely expensive. India has research ships that spend most of their time over the seas around India but that opportunity is yet to be exploited. To address this, an air-sea flux measurement system for operation on board research ships was planned. The system was tested on board Indian Research Vessels ORV SagarKanya during its cruise SK-296 in the Bay of Bengal (BoB) in July-August 2012, and NIO ship Sindhu Sadhana in June-July 2016. The complete set included instruments for measuring wind velocity, windspeed and direction, air and water temperature, humidity, pressure, all components of radiation and rainfall. In addition, ship motion was recorded at required sampling rate to correct for wind velocity. The set up facilitates the direct computation of sensible and latent heat fluxes using the eddy covariance method. In this thesis, design and installation of meteorological and ship motion sensors onboard research ships, data collection and quality control, computation of fluxes of heat, moisture and momentum using eddy covariance method and their comparison with those derived from bulk method are described. A set of sensors (hereafter, flux measuring system) were mounted on a retractable boom, ~7 m long forward of the bow to minimize the flow disturbance caused by the ship superstructures. The wind observed in the ship frame was corrected for ship motion contaminations. During the CTCZ cruise period true mean wind speed was over 10 m/s and true wind direction was South/South-Westerly. True windspeedis computed combiningdata from the anemometer a compass connected to AWS and a GPS. Turbulent fluxes were computed from motion-corrected time-series of high frequency velocity, water vapor, and air temperature data. Covariance latent heat flux, sensible heat flux, and wind stress were obtained by cross-correlating the motion-corrected vertical velocity with fast humidity fluctuations measured with anIR hygrometer, temperate fluctuation from sonic anemometer and motion-corrected horizontal windfluctuations from sonic anemometer, respectively. During the first attempt made in July-August 2012 as part of a cruise of CTCZ monsoonresearch program, observations were mainly taken in the North Bay of Bengal. The mean air-temperature and surface pressure were ~28 Deg C and ~998 hPa, respectively. Relative humidity was ~80%. Average wind speed varied in the range 4-12 m/s. The mean latent heat flux was 145 W/m2 , sensible heat flux was ~3 W/m2 and average sea-air temperature difference was ~ 0.7°C. The Bay of Bengal boundary layer experiment (BoBBLE) was conducted during June-July 2016 and the NIO research ship Sindhu Sadhana was deployed. The same suite of sensors installed during CTCZ were used during BoBBLE. During daytime, peaks of hourly net heat fluxes (Qnet ) were around 600 Wm-2(positive if into the sea), whereas, night time values were around -250 W m-2. Sea surface temperature was always >28°C and maximum air temperature exceeded 29°C. During the experimental period the mean Qnet was around -24 Wm-2 from both eddy covariance and conventional bulk methods, but there are significant differences on individual days.The new flux system gives fluxes which are superior to what was available before.