Publicação

Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose

Ver documento

Detalhes bibliográficos
Resumo:Digital Breast Tomosynthesis is a three-dimensional medical imaging technique that allows the view of sectional parts of the breast. Obtaining multiple slices of the breast constitutes an advantage in contrast to conventional mammography examination in view of the increased potential in breast cancer detectability. Conventional mammography, despite being a screening success, has undesirable specificity, sensitivity, and high recall rates owing to the overlapping of tissues. Although this new technique promises better diagnostic results, the acquisition methods and image reconstruction algorithms are still under research. Several articles suggest the use of analytic algorithms. However, more recent articles highlight the iterative algorithm’s potential for increasing image quality when compared to the former. The scope of this dissertation was to test the hypothesis of achieving higher quality images using iterative algorithms acquired with lower doses than those using analytic algorithms. In a first stage, the open-source Tomographic Iterative GPU-based Reconstruction (TIGRE) Toolbox for fast and accurate 3D x-ray image reconstruction was used to reconstruct the images acquired using an acrylic phantom. The algorithms used from the toolbox were the Feldkamp, Davis, and Kress, the Simultaneous Algebraic Reconstruction Technique, and the Maximum Likelihood Expectation Maximization algorithm. In a second and final state, the possibility of further reducing the radiation dose using image postprocessing tools was evaluated. A Total Variation Minimization filter was applied to the images reconstructed with the TIGRE toolbox algorithm that provided the best image quality. These were then compared to the images of the commercial unit used for the image acquisitions. With the use of image quality parameters, it was found that the Maximum Likelihood Expectation Maximization algorithm performance was the best of the three for lower radiation doses, especially with the filter. In sum, the result showed the potential of the algorithm in obtaining images with quality for low doses.
Autores principais:Diogo, Raquel Silva
Assunto:Tomossíntese Digital Mamária Feldkamp Davis and Kress Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados Maximum Likelihood Expectation Maximization Minimização da Variação Total dos Dados Teses de mestrado - 2023
Ano:2023
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Lisboa
Idioma:inglês
Origem:Repositório da Universidade de Lisboa
_version_ 1865920815160098816
author Diogo, Raquel Silva
author_facet Diogo, Raquel Silva
Diogo, Raquel Silva
author_role author
contributor_name_str_mv Matela, Nuno Miguel de Pinto Lobo e, 1978-
Repositório Científico de Acesso Aberto da ULisboa
country_str PT
creators_json_str [{\"Person.name\":\"Diogo, Raquel Silva\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Matela, Nuno Miguel de Pinto Lobo e, 1978-
Repositório Científico de Acesso Aberto da ULisboa
datacite.creators.creator.creatorName.fl_str_mv Diogo, Raquel Silva
datacite.date.Accepted.fl_str_mv 2023-01-01T00:00:00Z
datacite.date.available.fl_str_mv 2023-03-29T12:03:10Z
datacite.date.embargoed.fl_str_mv 2023-03-29T12:03:10Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
datacite.titles.title.fl_str_mv Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
dc.contributor.none.fl_str_mv Matela, Nuno Miguel de Pinto Lobo e, 1978-
Repositório Científico de Acesso Aberto da ULisboa
dc.creator.none.fl_str_mv Diogo, Raquel Silva
dc.date.Accepted.fl_str_mv 2023-01-01T00:00:00Z
dc.date.available.fl_str_mv 2023-03-29T12:03:10Z
dc.date.embargoed.fl_str_mv 2023-03-29T12:03:10Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10451/56892
dc.language.none.fl_str_mv eng
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
dc.title.fl_str_mv Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_bdcc
description Digital Breast Tomosynthesis is a three-dimensional medical imaging technique that allows the view of sectional parts of the breast. Obtaining multiple slices of the breast constitutes an advantage in contrast to conventional mammography examination in view of the increased potential in breast cancer detectability. Conventional mammography, despite being a screening success, has undesirable specificity, sensitivity, and high recall rates owing to the overlapping of tissues. Although this new technique promises better diagnostic results, the acquisition methods and image reconstruction algorithms are still under research. Several articles suggest the use of analytic algorithms. However, more recent articles highlight the iterative algorithm’s potential for increasing image quality when compared to the former. The scope of this dissertation was to test the hypothesis of achieving higher quality images using iterative algorithms acquired with lower doses than those using analytic algorithms. In a first stage, the open-source Tomographic Iterative GPU-based Reconstruction (TIGRE) Toolbox for fast and accurate 3D x-ray image reconstruction was used to reconstruct the images acquired using an acrylic phantom. The algorithms used from the toolbox were the Feldkamp, Davis, and Kress, the Simultaneous Algebraic Reconstruction Technique, and the Maximum Likelihood Expectation Maximization algorithm. In a second and final state, the possibility of further reducing the radiation dose using image postprocessing tools was evaluated. A Total Variation Minimization filter was applied to the images reconstructed with the TIGRE toolbox algorithm that provided the best image quality. These were then compared to the images of the commercial unit used for the image acquisitions. With the use of image quality parameters, it was found that the Maximum Likelihood Expectation Maximization algorithm performance was the best of the three for lower radiation doses, especially with the filter. In sum, the result showed the potential of the algorithm in obtaining images with quality for low doses.
dirty 0
eu_rights_str_mv openAccess
format masterThesis
fulltext.url.fl_str_mv https://repositorio.ulisboa.pt/bitstreams/f6f115af-ba66-40e5-8345-f07345d2a6bb/download
id ul_8bed5843f00a60ab583ec9bdee72ddc6
identifier.url.fl_str_mv http://hdl.handle.net/10451/56892
instacron_str ul
institution Universidade de Lisboa
instname_str Universidade de Lisboa
language eng
network_acronym_str ul
network_name_str Repositório da Universidade de Lisboa
oai_identifier_str oai:repositorio.ulisboa.pt:10451/56892
organization_str_mv urn:organizationAcronym:ul
person_str_mv Diogo, Raquel Silva
publishDate 2023
reponame_str Repositório da Universidade de Lisboa
repository_id_str urn:repositoryAcronym:ul
service_str_mv urn:repositoryAcronym:ul
spelling engpt_PTDigital Breast Tomosynthesis is a three-dimensional medical imaging technique that allows the view of sectional parts of the breast. Obtaining multiple slices of the breast constitutes an advantage in contrast to conventional mammography examination in view of the increased potential in breast cancer detectability. Conventional mammography, despite being a screening success, has undesirable specificity, sensitivity, and high recall rates owing to the overlapping of tissues. Although this new technique promises better diagnostic results, the acquisition methods and image reconstruction algorithms are still under research. Several articles suggest the use of analytic algorithms. However, more recent articles highlight the iterative algorithm’s potential for increasing image quality when compared to the former. The scope of this dissertation was to test the hypothesis of achieving higher quality images using iterative algorithms acquired with lower doses than those using analytic algorithms. In a first stage, the open-source Tomographic Iterative GPU-based Reconstruction (TIGRE) Toolbox for fast and accurate 3D x-ray image reconstruction was used to reconstruct the images acquired using an acrylic phantom. The algorithms used from the toolbox were the Feldkamp, Davis, and Kress, the Simultaneous Algebraic Reconstruction Technique, and the Maximum Likelihood Expectation Maximization algorithm. In a second and final state, the possibility of further reducing the radiation dose using image postprocessing tools was evaluated. A Total Variation Minimization filter was applied to the images reconstructed with the TIGRE toolbox algorithm that provided the best image quality. These were then compared to the images of the commercial unit used for the image acquisitions. With the use of image quality parameters, it was found that the Maximum Likelihood Expectation Maximization algorithm performance was the best of the three for lower radiation doses, especially with the filter. In sum, the result showed the potential of the algorithm in obtaining images with quality for low doses.application/pdfpt_PTComparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation doseDiogo, Raquel SilvaMatela, Nuno Miguel de Pinto Lobo e, 1978-HostingInstitutionOrganizationalRepositório Científico de Acesso Aberto da ULisboae-mailmailto:repositorio@reitoria.ulisboa.ptrepositorio@reitoria.ulisboa.ptURNurn:tid:2034936562023-03-29T12:03:10Z202320222023-01-01T00:00:00ZHandlehttp://hdl.handle.net/10451/56892http://purl.org/coar/access_right/c_abf2open accessTomossíntese Digital MamáriaFeldkamp Davis and KressSimultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos DadosMaximum Likelihood Expectation MaximizationMinimização da Variação Total dos DadosTeses de mestrado - 20232178050 bytesliteraturehttp://purl.org/coar/resource_type/c_bdccmaster thesishttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://repositorio.ulisboa.pt/bitstreams/f6f115af-ba66-40e5-8345-f07345d2a6bb/download
spellingShingle Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
Diogo, Raquel Silva
Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
Diogo, Raquel Silva
Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
status SINGLETON
subject.fl_str_mv Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
title Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
title_full Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
title_fullStr Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
title_full_unstemmed Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
title_short Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
title_sort Comparison of different image reconstruction algorithms for Digital Breast Tomosynthesis and assessment of their potential to reduce radiation dose
topic Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
topic_facet Tomossíntese Digital Mamária
Feldkamp Davis and Kress
Simultaneous Algebraic Reconstruction Technique; Maximum Likelihood Expectation Maximization; Minimização da Variação Total dos Dados
Maximum Likelihood Expectation Maximization
Minimização da Variação Total dos Dados
Teses de mestrado - 2023
url http://hdl.handle.net/10451/56892
visible 1