TY  - INPR
PB  - UCL (University College London)
N1  - Copyright © The Author 2020. Original content in this thesis is licensed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) Licence (https://creativecommons.org/licenses/by/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author?s request.
SP  - 1
TI  - Photoacoustic tomography: flexible acoustic solvers based on geometrical optics
EP  - 231
AV  - public
N2  - In this PhD project, we have developed acoustic solvers for 2D and 3D photoacoustic tomography (PAT). In the first part of the thesis, we review the literature related to the high frequency approximation to the solution of the wave equation. We include the principal families of methods and describe those more relevant from each family. Then we focus on ray tracing methods, which form the basis of our solver. We acknowledge that these methods have been thoroughly studied for seismic imaging and ocean acoustics. We split our contributions into theoretical and experimental. Theoretical. We further explore ray tracing methods and derive an approximation of the Green?s function based on the discretisation provided by ray trajectories. We describe the mathematical foundations for our forward and adjoint PAT acoustic operators. We include the limitations of our method and propose directions for future research to overcome them. Experimental. We provide extensive numerical experiments to test our solver. We start with 2D and 3D simulated data, using homogeneous and inhomogeneous sound speeds. We demonstrate the convergence for the homogeneous solver described in the theoretical results. We move then to 3D reconstructions with simulated data and experimental data. To obtain the reconstructions, we compare two families of optimisation algorithms (stochastic and deterministic), and we build a case for the use of stochastic methods when using our solver. We include the relevant background for these algorithms. In the last part of the project, we study Gaussian beams (GB). These are a natural extension of rays. We review GBs and propose forward and adjoint PAT operators based on a frequencial decomposition of the pressure waves.
Y1  - 2020/07/28/
ID  - discovery10104110
M1  - Doctoral
UR  - https://discovery-pp.ucl.ac.uk/id/eprint/10104110/
A1  - Rul·lan Palou De Comasema, Francisco
ER  -