Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics

Detalhes bibliográficos
Autor(a) principal: Dias Machado Costa, Jonatas
Data de Publicação: 2023
Outros Autores: Silva Santana, Lucas, Weber dos Santos, Rodrigo, Martins Rocha, Bernardo, de Oliveira Campos, Joventino
Tipo de documento: Artigo
Idioma: por
Título da fonte: Vetor (Online)
Texto Completo: https://periodicos.furg.br/vetor/article/view/16420
Resumo: The solution of linear systems plays a fundamental role in computer simulation software based on mathematical models to advance contemporary scientific research. Consequently, there is a growing demand for numerical methods and efficient implementations to face this challenge, particularly in the context of biomedical engineering where it is desired to use these simulators to create digital twins of patients and study certain pathological conditions. This work aims to explore and identify efficient techniques to solve linear systems related to the problem of cardiac biomechanics, thus accelerating simulations related to the intricate human cardiovascular system. To achieve this goal, several multigrid methods available in the AmgX library were selected, which were tested and analyzed in terms of their computational performance. As an initial step, problems based on Poisson's equation were solved considering simplified and complex geometries such as a cube and a human left ventricle. This study revealed distinct advantages associated with each method, depending on the complexity and format of the problems at hand.
id FURG-7_aa481b1a1d0eb2e1ecac1af18039596e
oai_identifier_str oai:ojs.periodicos.furg.br:article/16420
network_acronym_str FURG-7
network_name_str Vetor (Online)
repository_id_str
spelling Use of the AmgX Library for Solving Problems Related to Cardiac MechanicsAceleração da Solução Numérica de Problemas Da Biomecânica Cardíaca Utilizando Métodos Multigrid da Biblioteca AmgXComputational modelingLinear systemsMultigrid methodsGPUComputational modelingLinear systemsMultigrid methodsGPUModelagem computacionalSistemas linearesMétodos multigridGPUModelagem computacionalSistemas linearesMétodos multigridGPUThe solution of linear systems plays a fundamental role in computer simulation software based on mathematical models to advance contemporary scientific research. Consequently, there is a growing demand for numerical methods and efficient implementations to face this challenge, particularly in the context of biomedical engineering where it is desired to use these simulators to create digital twins of patients and study certain pathological conditions. This work aims to explore and identify efficient techniques to solve linear systems related to the problem of cardiac biomechanics, thus accelerating simulations related to the intricate human cardiovascular system. To achieve this goal, several multigrid methods available in the AmgX library were selected, which were tested and analyzed in terms of their computational performance. As an initial step, problems based on Poisson's equation were solved considering simplified and complex geometries such as a cube and a human left ventricle. This study revealed distinct advantages associated with each method, depending on the complexity and format of the problems at hand.A resolução de sistemas lineares desempenha um papel fundamental em softwares de simulações computacionais baseadas em modelos matemáticos para o avanço de pesquisas científicas contemporâneas. Consequentemente, há uma demanda crescente por métodos numéricos e implementações eficientes para enfrentar esse desafio, em particular no contexto da engenharia biomédica onde deseja-se utilizar esses simuladores para criar gêmeos digitais de pacientes e estudar determinadas condições patológicas. Este trabalho tem como objetivo explorar e identificar técnicas eficientes para resolver sistemas lineares relacionados ao problema da biomecânica cardíaca, acelerando assim as simulações relacionadas ao intrincado sistema cardiovascular humano. Para atingir esse objetivo, foram selecionados vários métodos multigrid disponíveis na biblioteca AmgX, que foram testados e analisados em termos do seu desempenho computacional. Como um passo inicial, problemas baseados na equação de Poisson, foram resolvidos considerando geometrias simplificadas e complexas como, por exemplo, um cubo e um ventrículo humano. Esse estudo revelou vantagens distintas associadas a cada método, dependendo da complexidade e do formato dos problemas em questão.Universidade Federal do Rio Grande2023-12-23info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://periodicos.furg.br/vetor/article/view/1642010.14295/vetor.v33i2.16420VETOR - Journal of Exact Sciences and Engineering; Vol. 33 No. 2 (2023); 32-40VETOR - Revista de Ciências Exatas e Engenharias; v. 33 n. 2 (2023); 32-402358-34520102-7352reponame:Vetor (Online)instname:Universidade Federal do Rio Grande (FURG)instacron:FURGporhttps://periodicos.furg.br/vetor/article/view/16420/10462Copyright (c) 2023 VETOR - Revista de Ciências Exatas e Engenhariasinfo:eu-repo/semantics/openAccessDias Machado Costa, JonatasSilva Santana, LucasWeber dos Santos, RodrigoMartins Rocha, Bernardode Oliveira Campos, Joventino2023-12-23T15:36:37Zoai:ojs.periodicos.furg.br:article/16420Revistahttps://periodicos.furg.br/vetorPUBhttps://periodicos.furg.br/vetor/oaigmplatt@furg.br2358-34520102-7352opendoar:2023-12-23T15:36:37Vetor (Online) - Universidade Federal do Rio Grande (FURG)false
dc.title.none.fl_str_mv Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
Aceleração da Solução Numérica de Problemas Da Biomecânica Cardíaca Utilizando Métodos Multigrid da Biblioteca AmgX
title Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
spellingShingle Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
Dias Machado Costa, Jonatas
Computational modeling
Linear systems
Multigrid methods
GPU
Computational modeling
Linear systems
Multigrid methods
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
title_short Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
title_full Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
title_fullStr Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
title_full_unstemmed Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
title_sort Use of the AmgX Library for Solving Problems Related to Cardiac Mechanics
author Dias Machado Costa, Jonatas
author_facet Dias Machado Costa, Jonatas
Silva Santana, Lucas
Weber dos Santos, Rodrigo
Martins Rocha, Bernardo
de Oliveira Campos, Joventino
author_role author
author2 Silva Santana, Lucas
Weber dos Santos, Rodrigo
Martins Rocha, Bernardo
de Oliveira Campos, Joventino
author2_role author
author
author
author
dc.contributor.author.fl_str_mv Dias Machado Costa, Jonatas
Silva Santana, Lucas
Weber dos Santos, Rodrigo
Martins Rocha, Bernardo
de Oliveira Campos, Joventino
dc.subject.por.fl_str_mv Computational modeling
Linear systems
Multigrid methods
GPU
Computational modeling
Linear systems
Multigrid methods
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
topic Computational modeling
Linear systems
Multigrid methods
GPU
Computational modeling
Linear systems
Multigrid methods
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
Modelagem computacional
Sistemas lineares
Métodos multigrid
GPU
description The solution of linear systems plays a fundamental role in computer simulation software based on mathematical models to advance contemporary scientific research. Consequently, there is a growing demand for numerical methods and efficient implementations to face this challenge, particularly in the context of biomedical engineering where it is desired to use these simulators to create digital twins of patients and study certain pathological conditions. This work aims to explore and identify efficient techniques to solve linear systems related to the problem of cardiac biomechanics, thus accelerating simulations related to the intricate human cardiovascular system. To achieve this goal, several multigrid methods available in the AmgX library were selected, which were tested and analyzed in terms of their computational performance. As an initial step, problems based on Poisson's equation were solved considering simplified and complex geometries such as a cube and a human left ventricle. This study revealed distinct advantages associated with each method, depending on the complexity and format of the problems at hand.
publishDate 2023
dc.date.none.fl_str_mv 2023-12-23
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://periodicos.furg.br/vetor/article/view/16420
10.14295/vetor.v33i2.16420
url https://periodicos.furg.br/vetor/article/view/16420
identifier_str_mv 10.14295/vetor.v33i2.16420
dc.language.iso.fl_str_mv por
language por
dc.relation.none.fl_str_mv https://periodicos.furg.br/vetor/article/view/16420/10462
dc.rights.driver.fl_str_mv Copyright (c) 2023 VETOR - Revista de Ciências Exatas e Engenharias
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2023 VETOR - Revista de Ciências Exatas e Engenharias
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade Federal do Rio Grande
publisher.none.fl_str_mv Universidade Federal do Rio Grande
dc.source.none.fl_str_mv VETOR - Journal of Exact Sciences and Engineering; Vol. 33 No. 2 (2023); 32-40
VETOR - Revista de Ciências Exatas e Engenharias; v. 33 n. 2 (2023); 32-40
2358-3452
0102-7352
reponame:Vetor (Online)
instname:Universidade Federal do Rio Grande (FURG)
instacron:FURG
instname_str Universidade Federal do Rio Grande (FURG)
instacron_str FURG
institution FURG
reponame_str Vetor (Online)
collection Vetor (Online)
repository.name.fl_str_mv Vetor (Online) - Universidade Federal do Rio Grande (FURG)
repository.mail.fl_str_mv gmplatt@furg.br
_version_ 1797041760293093376

Registros relacionados