EVENTO
Gaussian Process Modeling with Applications to Tumor Growth
Tipo de evento: Defesa de Dissertação de Mestrado
Mechanistic tumor growth models have been used to provide a better understanding of how the disease evolves and help the development of therapies. Aside from simulating a certain tumor growth model, it is fundamental to infer its parameters using available experimental data. This process incurs in the solution of an inverse problem, which requires repeated model evaluation, particularly using Bayesian inference. This may be computationally prohibitive for complex models, speciallymultiscale models. There are techniques of performing approximate Bayesian inference, which reducethe overall computational time, although these may still be infeasible if the model is expensive tobe evaluated. In order to alleviate the difficulty in this procedure, we propose the use of a surrogate (or metamodel). Our surrogate model is a Gaussian Process, a data-driven model recently used in machine learning and metamodeling context. This work reviews Gaussian processes (GP) theoretical and practical aspects in a regression problem, discussing how to construct an adequate GP for a particular problem, covering recent developments in this area. We then define a GP surrogate modeland combine it with an approximate bayesian computation method, the ABC-MCMC method for solving the inverse problem. We compare the standard and proposed approaches in two tumor growth models. Results suggest that the strategy is promising in terms of reducing the computational cost. Moreover, we comment on how to adaptively construct the surrogate, so as to improve even further the overall process efficiency.Para assistir acesse: https://us02web.zoom.us/j/84833502639?pwd=M1kwejZ3TVdna2tac0xmZmxzckt0dz09
Data Início: 05/08/2021 Hora: 13:30 Data Fim: 05/08/2021 Hora: 17:00
Local: LNCC - Laboratório Nacional de Computação Ciêntifica - Webinar
Data Início: 05/08/2021 Hora: 14:00 Data Fim: 05/08/2021 Hora: 17:00
Local: LNCC - Laboratório Nacional de Computação Ciêntifica - Webinar
Aluno: Henrique Matheus Ferreira da Silva - - LNCC
Orientador: Fabio André Machado Porto - Laboratório Nacional de Computação Científica - LNCC
Participante Banca Examinadora: Eduardo Bezerra da Silva - CEFET - RJ Fabio André Machado Porto - Laboratório Nacional de Computação Científica - LNCC Marisa Fabiana Nicolás - Laboratório Nacional de Computação Científica - LNCC
