Dans le domaine de l'imagerie ultrasonore, l'imagerie non linéaire est devenue une branche importante. Imagerie non linéaire peut être divisée en imagerie harmonique tissulaire et imagerie harmonique de contraste, selon l'endroit où les signaux non linéaires viennent. Imagerie harmonique de contraste apparaît parce que les agents de contraste, qui sont injectés par voie intraveineuse pour améliorer les faibles échos rétrodiffusés par des cellules sanguines, peuvent vibrer non linéaire quand ils subissent une pression acoustique. Ces signaux non linéaires rétrodiffusés par des agents de contraste sont utilisés pour former des images harmoniques. Toutefois, lors de la propagation des ondes dans les tissus, les harmoniques de l'onde sont également générés dans le tissu. La présence de signaux harmoniques de tissus dégrade la qualité d'image en contraste imagerie harmonique. Cette thèse vise à mieux distinguer les échos des agents de contraste et les échos de tissus, concevant des nouvelles modalités, ou optimisant les modalités existantes. Nos efforts se concentrent principalement sur les techniques multi-impulsions en imagerie ultrasonore de contraste. Tout d'abord, nous proposons une formule de généraliser la plupart des techniques multi- impulsions. La formulation peut être utilisée pour prédire les éléments non linéaires dans chaque bande fréquentiel et de concevoir de nouvelles séquences de transmission pour augmenter ou diminuer composants non linéaires. Les résultats des simulations sur plusieurs techniques multi- impulsions sont en accord avec les résultats donnés dans les littératures précédentes. Deuxièmement, les techniques utilisant multi transmissions pour augmenter le CTR sont généralement basés sur la réponse des diffuseurs statiques. Cependant, diffuseur en mouvement a une influence inévitable sur des techniques. Il peut améliorer ou dégrader la technique. Des simulations, des expériences in vitro à partir d'une seule bulle et les nuages de bulles, et des expériences sur des rats montrent que le déphasage des échos rétrodiffusés par des bulles dépend du déphasage des transmissions, et que le mouvement de la bulle influe sur l’efficacité des techniques multi- impulsions. En outre, les résultats expérimentaux basés sur la technique de l'inversion de seconde harmonique (SHI) révèlent que le mouvement de la bulle peut être pris en compte pour optimiser les techniques multi-impulsions. Par ailleurs, une nouvelle technique, appelée double inversion de impulsion (DPI), a également été proposé. La technique PI est appliquée deux fois avant et après l'arrivée de l'agent de contraste dans la région d'intérêt. Les signaux résultants sont soustraits pour supprimer les harmoniques du tissu. Les simulations et les résultats in vitro ont montré une amélioration de CTR de DPI. Toutefois, la présence de mouvements de tissus peut limiter à l'efficacité de cette technique. Résultats in vivo confirment cette limitation. / In ultrasound imaging domain, nonlinear imaging has become an important branch. Nonlinear imaging can be divided into tissue harmonic imaging and contrast harmonic imaging, according to where the nonlinear signals come from. Contrast harmonic imaging emerges because contrast agents, which are intravenously injected to enhance the weak echoes backscattered from blood cells, can vibrate nonlinearly when they undergo an acoustic pressure. Then, these nonlinear signals backscattered by contrast agents are collected to form harmonic images. However, during the wave propagation in tissue, the harmonics of the transmitted wave are also generated in tissue. The presence of tissue harmonic signals degrades the image quality in contrast harmonic imaging. This thesis aims to better distinguish the echoes from contrast agents and the echoes from tissue, whether through designing new modalities, or investigating and optimizing the existing modalities. Our efforts are mainly focused on the multi-pulse techniques in ultrasound contrast imaging. Firstly, we propose a mathematical background to generalize most of the multi-pulse ultrasound imaging techniques that have been described in previous literatures. The formulation can be used to predict the nonlinear components in each frequency band and to design new transmission sequences to either increase or decrease specified nonlinear components in each harmonic band. Simulation results on several multi-pulse techniques are in agreement with the results given in previous literatures. Secondly, the techniques using multiple transmissions to increase the CTR are generally based on the response of static scatterers inside the imaged region. However, scatterer motion, for example in blood vessels, has an inevitable influence on the relevance of the techniques. It can either upgrade or degrade the technique involved. Simulations, in-vitro experiments from a single bubble and clouds of bubbles, and in-vivo experiments from rats show that the phase shift of the echoes backscattered from bubbles is dependent on the transmissions’ phase shift, and that the bubble motion influences the efficiency of multi-pulse techniques. Furthermore, experimental results based on the second-harmonic inversion (SHI) technique reveal that bubble motion can be taken into account to optimize multi-pulse techniques. Besides, a new technique, called double pulse inversion (DPI), has also been proposed. The PI technique is applied twice before and after the arrival of the contrast agents to the region of interest. The resulting PI signals are substracted to suppress the tissue-generated harmonics and to improve CTR. Simulations and in-vitro experimental results have shown an improved CTR of DPI. However, the presence of tissue movements may hamper the effectiveness of this technique. In-vivo experimental results confirm that the tissue motion of the rat during the acquisition is an inevitable barrier of this technique.

In ultrasound imaging domain, nonlinear imaging has become an important branch. Nonlinear imaging can be divided into tissue harmonic imaging and contrast harmonic imaging, according to where the nonlinear signals come from. Contrast harmonic imaging emerges because contrast agents, which are intravenously injected to enhance the weak echoes backscattered from blood cells, can vibrate nonlinearly when they undergo an acoustic pressure. Then, these nonlinear signals backscattered by contrast agents are collected to form harmonic images. However, during the wave propagation in tissue, the harmonics of the transmitted wave are also generated in tissue. The presence of tissue harmonic signals degrades the image quality in contrast harmonic imaging. This thesis aims to better distinguish the echoes from contrast agents and the echoes from tissue, whether through designing new modalities, or investigating and optimizing the existing modalities. Our efforts are mainly focused on the multi-pulse techniques in ultrasound contrast imaging. Firstly, we propose a mathematical background to generalize most of the multi-pulse ultrasound imaging techniques that have been described in previous literatures. The formulation can be used to predict the nonlinear components in each frequency band and to design new transmission sequences to either increase or decrease specified nonlinear components in each harmonic band. Simulation results on several multi-pulse techniques are in agreement with the results given in previous literatures. Secondly, the techniques using multiple transmissions to increase the CTR are generally based on the response of static scatterers inside the imaged region. However, scatterer motion, for example in blood vessels, has an inevitable influence on the relevance of the techniques. It can either upgrade or degrade the technique involved. Simulations, in-vitro experiments from a single bubble and clouds of bubbles, and in-vivo experiments from rats show that the phase shift of the echoes backscattered from bubbles is dependent on the transmissions' phase shift, and that the bubble motion influences the efficiency of multi-pulse techniques. Furthermore, experimental results based on the second-harmonic inversion (SHI) technique reveal that bubble motion can be taken into account to optimize multi-pulse techniques. Besides, a new technique, called double pulse inversion (DPI), has also been proposed. The PI technique is applied twice before and after the arrival of the contrast agents to the region of interest. The resulting PI signals are substracted to suppress the tissue-generated harmonics and to improve CTR. Simulations and in-vitro experimental results have shown an improved CTR of DPI. However, the presence of tissue movements may hamper the effectiveness of this technique. In-vivo experimental results confirm that the tissue motion of the rat during the acquisition is an inevitable barrier of this technique.

This thesis deals with the principles of ultrasonic perfusion analysis and methods for determining perfusion parameters. It examines Evolutionary algorithms and their ability to optimize the approximation of dilution curves from ultrasond tissue scannig. It compares the optimization performance of three evolutionary algorithms. Continuous genetic algorithm GA, algorithm SOMA and PSO. Methods are evaluated on simulated and clinical data.