Optimizing a medical image registration algorithm based on profiling data for real-time performance
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1007/s11042-021-11699-x http://hdl.handle.net/11449/222782 |
Resumo: | Image registration is a commonly task in medical image analysis. Therefore, a significant number of algorithms have been developed to perform rigid and non-rigid image registration. Particularly, the free-form deformation algorithm is frequently used to carry out non-rigid registration task; however, it is a computationally very intensive algorithm. In this work, we describe an approach based on profiling data to identify potential parts of this algorithm for which parallel implementations can be developed. The proposed approach assesses the efficient of the algorithm by applying performance analysis techniques commonly available in traditional computer operating systems. Hence, this article provides guidelines to support researchers working on medical image processing and analysis to achieve real-time non-rigid image registration applications using common computing systems. According to our experimental findings, significant speedups can be accomplished by parallelizing sequential snippets, i.e., code regions that are executed more than once. For the selected costly functions previously identified in the studied free-form deformation algorithm, the developed parallelization decreased the runtime by up to seven times relatively to the related single thread based implementation. The implementations were developed based on the Open Multi-Processing application programming interface. In conclusion, this study confirms that based on the call graph visualization and detected performance bottlenecks, one can easily find and evaluate snippets which are potential optimization targets in addition to throughput in memory accesses. |
id |
UNSP_2a1e35fa0768051f07875f872b0d0f27 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/222782 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Optimizing a medical image registration algorithm based on profiling data for real-time performanceMedical image processing and analysisNon-rigid image registrationPerformance analysisProfiling toolsImage registration is a commonly task in medical image analysis. Therefore, a significant number of algorithms have been developed to perform rigid and non-rigid image registration. Particularly, the free-form deformation algorithm is frequently used to carry out non-rigid registration task; however, it is a computationally very intensive algorithm. In this work, we describe an approach based on profiling data to identify potential parts of this algorithm for which parallel implementations can be developed. The proposed approach assesses the efficient of the algorithm by applying performance analysis techniques commonly available in traditional computer operating systems. Hence, this article provides guidelines to support researchers working on medical image processing and analysis to achieve real-time non-rigid image registration applications using common computing systems. According to our experimental findings, significant speedups can be accomplished by parallelizing sequential snippets, i.e., code regions that are executed more than once. For the selected costly functions previously identified in the studied free-form deformation algorithm, the developed parallelization decreased the runtime by up to seven times relatively to the related single thread based implementation. The implementations were developed based on the Open Multi-Processing application programming interface. In conclusion, this study confirms that based on the call graph visualization and detected performance bottlenecks, one can easily find and evaluate snippets which are potential optimization targets in addition to throughput in memory accesses.Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)CNPq National Scientific and Technological Development Council Research Group PIXEL - UNEMATPrograma Doutoral em Engenharia Informática Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial Faculdade de Engenharia Universidade do PortoDepartamento de Ciências da Computação Faculdade de Ciências Universidade Estadual Paulista-UNESPInstituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial Departamento de Engenharia Mecânica Faculdade de Engenharia Universidade do PortoDepartamento de Ciências da Computação Faculdade de Ciências Universidade Estadual Paulista-UNESPResearch Group PIXEL - UNEMATUniversidade do PortoUniversidade Estadual Paulista (UNESP)Gulo, Carlos A. S. J.Sementille, Antonio C. [UNESP]Tavares, João Manuel R. S.2022-04-28T19:46:39Z2022-04-28T19:46:39Z2022-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article2603-2620http://dx.doi.org/10.1007/s11042-021-11699-xMultimedia Tools and Applications, v. 81, n. 2, p. 2603-2620, 2022.1573-77211380-7501http://hdl.handle.net/11449/22278210.1007/s11042-021-11699-x2-s2.0-85118383046Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengMultimedia Tools and Applicationsinfo:eu-repo/semantics/openAccess2022-04-28T19:46:39Zoai:repositorio.unesp.br:11449/222782Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462022-04-28T19:46:39Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
title |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
spellingShingle |
Optimizing a medical image registration algorithm based on profiling data for real-time performance Gulo, Carlos A. S. J. Medical image processing and analysis Non-rigid image registration Performance analysis Profiling tools |
title_short |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
title_full |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
title_fullStr |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
title_full_unstemmed |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
title_sort |
Optimizing a medical image registration algorithm based on profiling data for real-time performance |
author |
Gulo, Carlos A. S. J. |
author_facet |
Gulo, Carlos A. S. J. Sementille, Antonio C. [UNESP] Tavares, João Manuel R. S. |
author_role |
author |
author2 |
Sementille, Antonio C. [UNESP] Tavares, João Manuel R. S. |
author2_role |
author author |
dc.contributor.none.fl_str_mv |
Research Group PIXEL - UNEMAT Universidade do Porto Universidade Estadual Paulista (UNESP) |
dc.contributor.author.fl_str_mv |
Gulo, Carlos A. S. J. Sementille, Antonio C. [UNESP] Tavares, João Manuel R. S. |
dc.subject.por.fl_str_mv |
Medical image processing and analysis Non-rigid image registration Performance analysis Profiling tools |
topic |
Medical image processing and analysis Non-rigid image registration Performance analysis Profiling tools |
description |
Image registration is a commonly task in medical image analysis. Therefore, a significant number of algorithms have been developed to perform rigid and non-rigid image registration. Particularly, the free-form deformation algorithm is frequently used to carry out non-rigid registration task; however, it is a computationally very intensive algorithm. In this work, we describe an approach based on profiling data to identify potential parts of this algorithm for which parallel implementations can be developed. The proposed approach assesses the efficient of the algorithm by applying performance analysis techniques commonly available in traditional computer operating systems. Hence, this article provides guidelines to support researchers working on medical image processing and analysis to achieve real-time non-rigid image registration applications using common computing systems. According to our experimental findings, significant speedups can be accomplished by parallelizing sequential snippets, i.e., code regions that are executed more than once. For the selected costly functions previously identified in the studied free-form deformation algorithm, the developed parallelization decreased the runtime by up to seven times relatively to the related single thread based implementation. The implementations were developed based on the Open Multi-Processing application programming interface. In conclusion, this study confirms that based on the call graph visualization and detected performance bottlenecks, one can easily find and evaluate snippets which are potential optimization targets in addition to throughput in memory accesses. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-04-28T19:46:39Z 2022-04-28T19:46:39Z 2022-01-01 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1007/s11042-021-11699-x Multimedia Tools and Applications, v. 81, n. 2, p. 2603-2620, 2022. 1573-7721 1380-7501 http://hdl.handle.net/11449/222782 10.1007/s11042-021-11699-x 2-s2.0-85118383046 |
url |
http://dx.doi.org/10.1007/s11042-021-11699-x http://hdl.handle.net/11449/222782 |
identifier_str_mv |
Multimedia Tools and Applications, v. 81, n. 2, p. 2603-2620, 2022. 1573-7721 1380-7501 10.1007/s11042-021-11699-x 2-s2.0-85118383046 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Multimedia Tools and Applications |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
2603-2620 |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799964697336741888 |