Devant l'impact des maladies neurodégénératives sur la société, les thérapies par ultrasons focalisés apparaissent comme des techniques prometteuses combinant non invasivité, précision spatiale millimétrique et capacité d'atteindre les structures profondes du cerveau. Cependant, des travaux sont encore nécessaires pour renforcer les effets de la neuromodulation, comprendre les mécanismes sous-jacents et contrôler la sûreté de la technique avant d'entreprendre des essais cliniques. Dans cette thèse, la propagation des ultrasons dans le cerveau de rongeurs et de singes a été étudiée numériquement afin d'estimer l'intensité acoustique dans le cerveau, la répartition spatiale des ondes dans la boîte crânienne et l'élévation de température. Afin d'évaluer physiologiquement les effets des ultrasons à l'échelle cellulaire, l'activité de neurones uniques a été mesurée sur des macaques éveillés pendant une neuromodulation ultrasonore. Puis, la durée de l’effet de modulation a été augmentée grâce à une prolongation du tir sur des singes exécutant une tâche visuelle. L'imagerie fonctionnelle par IRM a ensuite permis de faire ressortir des changements de connectivité entre l'aire stimulée et des régions du cerveau éloignées.Enfin, les avantages de la neurostimulation par ultrasons ont été combinés avec l'efficacité d'un agent neuroactif. En utilisant des microbulles conjointement aux ultrasons, la barrière hémato-encéphalique a été ouverte localement et réversiblement dans le cortex visuel de macaques anesthésiés pour permettre le passage d'un neurotransmetteur inhibiteur dans le cerveau. La baisse d'amplitude des réponses EEG du cortex visuel à des stimuli démontre la faisabilité de la délivrance locale et non invasive de neuromodulateurs dans le cerveau. Ainsi, les paramètres ultrasonores ont été optimisés grâce aux simulations numériques et à des expériences in vivo pour renforcer les effets de neuromodulation tout en contrôlant les effets indésirables, avec l'objectif de se diriger vers des applications thérapeutiques et proposer de nouveaux outils pour des études de connectivité cérébrale / Considering the extent of neurodegenerative diseases consequences on the society, focused ultrasound appears as a promising technique combining non-invasiveness, millimetric spatial accuracy and ability to reach deep brain structures. However, efforts still need to be made to amplify the effects of focused ultrasound neuromodulation, understand its mechanism and control the safety of the technique before moving towards human trials.The ultrasound propagation inside the brain of rodents and monkeys was first studied numerically to estimate the maximum intensity in the brain, the pressure distribution in the skull cavity and the thermal rise. To evaluate physiologically the ultrasound effects at the cellular level, the activity of individual neurons was measured on awake macaques during ultrasonic neuromodulation. To further increase the duration of the modulation, a longer sonication was tested successfully on macaques performing a visual task. Functional MRI was then used to highlight the connectivity changes between the stimulated area and distant cerebral regions. Finally, the advantages of ultrasound neurostimulation were combined with the efficiency of a neuroactive agent (GABA). Using microbubbles and ultrasound, the blood brain barrier was opened locally and reversibly in the visual cortex of anesthetized macaques to deliver an inhibitory neurotransmitter in the brain. The amplitude of the EEG response of the visual cortex to stimuli decreased after GABA injection, demonstrating the feasibility of delivering neuroactive drugs non-invasively and locally to any brain region.Overall, ultrasound parameters were optimized with both numerical tools and in vivo experiments to amplify neuromodulation effects while controlling the safety. This work opens the way to the development of novel therapeutic applications and new tools for connectivity studies

It is estimated that mind-boggling 14.1 million new cases of cancer occurred worldwide in 2012 alone. This number is alarming. Although healthy lifestyle may reduce a risk of developing cancer, there is always some probability that cancer would develop even in an absolutely fit individual. There are two main conditions for successful treatment of cancer. Firstly, early diagnostic is absolutely crucial. Secondly, there is a need for suitable surgical methods for affected tissue removal. Ultrasound has a great potential to be used for both purposes as a non-invasive method. Photoacoustic spectroscopy is imaging method for tumor detection of great properties making the use of ultrasound while High-Intensity Focused Ultrasound (HIFU) is non-invasive surgical method. These methods would be impossible without precise ultrasound propagation simulations. The k-Wave is an open source MATLAB toolbox implementing such simulations. So, why are not these methods already deployed in treatment? Unfortunately, the simulation of ultrasound propagation is a very time consuming task, which makes it ineffective for medical purposes. However, there are a few options how to accelerate these simulations. The use of GPU is a very promising way to accelerate simulation.   The main topic of this thesis is the acceleration of the simulation of soundwaves propagation in bones and hard tissue. The implementation developed as a part of this thesis was benchmarked on various supercomputers including Anselm in Ostrava and Piz Daint in Lugano. The implemented solution provides remarkable acceleration compared to the original MATLAB prototype. It was able to accelerate the simulation around 160 times in the best case. It means that the simulation, which would otherwise last for 6.5 days, can be now computed in one hour. This acceleration was achieved using an NVIDIA Tesla P100 to run the simulation with the domain size of 416x416x416 grid points. The thesis includes performance benchmarks on different GPUs to provide complex image acceleration capabilities of developed implementation and provides discussion about memory usage and numerical accuracy. Thanks to the implemented solution harnessing the power of modern GPUs, doctors and researchers all around the world have a powerful tool in hands.

After the introduction of CUDA by Nvidia, the GPUs became devices capable of accelerating any general purpose computation. GPUs are designed as parallel processors which posses huge computation power. Modern supercomputers are often equipped with GPU accelerators. Sometimes the performance or the memory capacity of a single GPU is not enough for a scientific application. The application needs to be scaled into multiple GPUs. During the computation there is need for the GPUs to exchange partial results. This communication represents computation overhead. For this reason it is important to research the methods of the effective communication between GPUs. This means less CPU involvement, lower latency, shared system buffers. Inter-node and intra-node communication is examined. The main focus is on GPUDirect technologies from Nvidia and CUDA-Aware MPI. Subsequently k-Wave toolbox for simulating the propagation of acoustic waves is introduced. This application is accelerated by using CUDA-Aware MPI.

This thesis deals with an implementation of non-blocking I/O interface for the k-Wave project, which is designed for time-domain simulation of ultrasound propagation. Main focus is on large domain simulations that, due to high computing power requirements, must run on supercomputers and produce tens of GB of data in a single simulation step. In this thesis, I have designed and implemented a non-blocking interface for storing data using dedicated threads, which allows to overlap simulation calculations with disk operations in order to speed up the simulation. An acceleration of up to 33% was achieved compared to the current implementation of project k-Wave, which resulted, among other things, also to reduce cost of the simulation.

This thesis explores potential use of GPGPU libraries to accelerate k-Wave toolkit's acoustic wave propagation simulation. Firstly, the thesis researches and assesses available high level GPGPU libraries. Afterwards, an insight into k-Wave toolkit's current state of simulation acceleration is provided. Based on that, an approach to enhance currently available code for processors into a heterogeneous application, that is capable of being run on graphics card, is proposed. The outcome of this thesis is an application that can utilize graphics card. If graphics card is unavailable, a fallback into thread and SIMD based acceleration for processor is executed. The product of this thesis is then evaluated based on its performance, maintenance difficulty and usability.

This work is part of the k-Wave project, which is a toolbox designed for time ultrasound simulations in complex and heterogeneous media. The simulation functions are based on the k-space pseudospectral method. The goal of this work is to compute these simulations on graphics cards using local domain decompostion. Thanks to decomposition we could compute these simulations faster, and on larger data grids. The main goal of this work is to achieve efficiency and scalability.

After the introduction of CUDA by Nvidia, the GPUs became devices capable of accelerating any general purpose computation. GPUs are designed as parallel processors which posses huge computation power. Modern supercomputers are often equipped with GPU accelerators. Sometimes single GPU performance is not enough for a scientific application and it needs to scale over multiple GPUs. During the computation, there is a need for the GPUs to exchange partial results. This communication represents computation overhead and it is important to research methods of the effective communication between GPUs. This means less CPU involvement, lower latency and shared system buffers. This thesis is focused on inter-node and intra-node GPU-to-GPU communication using GPUDirect technologies from Nvidia and CUDA-Aware MPI. Subsequently, k-Wave toolbox for simulating the propagation of acoustic waves is introduced. This application is accelerated by using CUDA-Aware MPI. Peer-to-peer transfer support is also integrated to k-Wave using CUDA Inter-process Communication.

This work deals with the 2D and 3D visualization of simulation outputs from the k-Wave toolbox. This toolbox, designed to accurately model the propagation of ultrasound waves in the human body, usually generates immense amounts of output data (up to hundreds of GB). That is why new methods for both the visualization and the effective data representation are necessary to be developed to help users to easily understand the simulation results. This thesis elaborates on the data format, simulation outputs are stored in, with the use of the HDF5 library and looking for the best way to quickly read the simulation data. Finally, the thesis presents the design and the implementation of the console-based application for big simulation data pre-processing and the GUI-based application for interactive visualization of the pre-processed data. The most significant features of these applications are downsampling data, changing the format of storing, viewing 2D sections, planar and volumetric visualization and animation of the simulation process. The proposed implementation allows parts of the simulation domain to be visualised within tens of milliseconds even if the simulation domain comprises GBs of data - This significantly streamlines the work of scientists and clinicians in the field of HIFU.

This thesis deals with an effective solution of the parallel I/O of the k-Wave tool, which is designed for time domain acoustic and ultrasound simulations. k-Wave is a supercomputer application, it runs on a Lustre file system and it requires to be implemented with MPI and stores the data in suitable data format (HDF5). I designed three methods of optimization which fits k-Wave's needs. It uses accumulation and redistribution techniques. In comparison with the native write, every optimization method led to better write speed, up to 13.6GB/s. It is possible to use these methods to optimize every data distributed application with the write speed issue.

Nowadays, the simulation of ultrasound acoustic waves has a wide range of practical usage. As one of them we can name the simulation in realistic tissue media, which is successfully used in medicine. There are several software applications dedicated to perform such simulations. k-Wave is one of them. The computational difficulty of the simulation itself is very high, and this leaves a space to explore new speed-up methods. In this master's thesis, we proposed a way to speed-up the simulation based on parallelization using Intel Xeon Phi accelerator. The accelerator contains large amount of cores and an extra-wide vector unit, and therefore, is ideal for purpose of parallelization and vectorization. The implementation is using OpenMP version 4.0, which brings some new options such as explicit vectorization. Results were measured during extensive experiments.