Publicação

Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study

Ver documento

Detalhes bibliográficos
Resumo:ABSTRACT - Myocardial perfusion imaging (MPI) by single-photon emission computed tomography (SPECT) plays a crucial role in the diagnosis of coronary artery disease. Moreover, the quantification of these images typically involves the extraction of quantitative parameters obtained from the rest stress perfusion. However, the acquisition systems have some limitations such as spatial blurring and low-count data, which may introduce bias in the classification. Additionally, these processes are time-consuming and user-dependent, leading to significant intra and inter-operator variability. Furthermore, over the years there has been a constant effort to reduce the dose of MPI. In this sense, both the variability classification issues and the dose reduction concerns can impact the true assessment of SPECT-MPI. In recent years, with the rise of artificial intelligence algorithms, several studies have proposed automatic Deep Learning techniques for the classification of MPI, moreover regarding low-count data. In this project, we ran 5 Convolutional Neural Network models with pre-trained weights: one trained on stress real full-time data (100%, as 100R), three individual models with synthetic 75%, 50%, and 25% count settings, and another one with all datasets combined (ALL). Thus, we compared their performance when tested on full-time and low-time studies and assessed the application of synthetic subsampled data from the Poisson Resampling technique in SPECT-MPI classification tasks. In conclusion, both 100R and ALL models achieved good and similar results when tested in real full-time (the 100R model achieved an accuracy of 0.70 and the ALL model achieved an accuracy of 0.65) and real low-time at 75% (both models achieved an accuracy of 0.71). Bellow this percentage, the models’ accuracy began to drop, possibly due to the limited information these images contain. Thus, subsampled data from a Poisson resampling method may be a possible solution to conduct further studies regarding the classification of low-time SPECT-MPI.
Autores principais:Gonçalves, Gonçalo Cravo de Jesus
Assunto:Deep learning Convolutional neural networks Image classification SPECT Myocardial perfusion imaging Nuclear medicine Aprendizagem profunda Redes neuronais convolucionais Classificação de imagens SPECT Cintigrafia de perfusão do miocárdio Medicina nuclear MEB
Ano:2023
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Instituto Politécnico de Lisboa
Idioma:inglês
Origem:Repositório Científico do Instituto Politécnico de Lisboa
_version_ 1863853063823949824
author Gonçalves, Gonçalo Cravo de Jesus
author_facet Gonçalves, Gonçalo Cravo de Jesus
author_role author
contributor_name_str_mv Figueiredo, Sérgio
Jorge, Pedro Miguel Torres Mendes
RCIPL
country_str PT
creators_json_str [{\"Person.name\":\"Gonçalves, Gonçalo Cravo de Jesus\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Figueiredo, Sérgio
Jorge, Pedro Miguel Torres Mendes
RCIPL
datacite.creators.creator.creatorName.fl_str_mv Gonçalves, Gonçalo Cravo de Jesus
datacite.date.Accepted.fl_str_mv 2023-12-01T00:00:00Z
datacite.date.available.fl_str_mv 2024-08-28T09:29:27Z
datacite.date.embargoed.fl_str_mv 2024-08-28T09:29:27Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
datacite.titles.title.fl_str_mv Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
dc.contributor.none.fl_str_mv Figueiredo, Sérgio
Jorge, Pedro Miguel Torres Mendes
RCIPL
dc.creator.none.fl_str_mv Gonçalves, Gonçalo Cravo de Jesus
dc.date.Accepted.fl_str_mv 2023-12-01T00:00:00Z
dc.date.available.fl_str_mv 2024-08-28T09:29:27Z
dc.date.embargoed.fl_str_mv 2024-08-28T09:29:27Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10400.21/17628
dc.language.none.fl_str_mv eng
dc.publisher.none.fl_str_mv Instituto Politécnico de Lisboa, Escola Superior de Tecnologia da Saúde de Lisboa
dc.rights.cclincense.fl_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
dc.title.fl_str_mv Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_bdcc
description ABSTRACT - Myocardial perfusion imaging (MPI) by single-photon emission computed tomography (SPECT) plays a crucial role in the diagnosis of coronary artery disease. Moreover, the quantification of these images typically involves the extraction of quantitative parameters obtained from the rest stress perfusion. However, the acquisition systems have some limitations such as spatial blurring and low-count data, which may introduce bias in the classification. Additionally, these processes are time-consuming and user-dependent, leading to significant intra and inter-operator variability. Furthermore, over the years there has been a constant effort to reduce the dose of MPI. In this sense, both the variability classification issues and the dose reduction concerns can impact the true assessment of SPECT-MPI. In recent years, with the rise of artificial intelligence algorithms, several studies have proposed automatic Deep Learning techniques for the classification of MPI, moreover regarding low-count data. In this project, we ran 5 Convolutional Neural Network models with pre-trained weights: one trained on stress real full-time data (100%, as 100R), three individual models with synthetic 75%, 50%, and 25% count settings, and another one with all datasets combined (ALL). Thus, we compared their performance when tested on full-time and low-time studies and assessed the application of synthetic subsampled data from the Poisson Resampling technique in SPECT-MPI classification tasks. In conclusion, both 100R and ALL models achieved good and similar results when tested in real full-time (the 100R model achieved an accuracy of 0.70 and the ALL model achieved an accuracy of 0.65) and real low-time at 75% (both models achieved an accuracy of 0.71). Bellow this percentage, the models’ accuracy began to drop, possibly due to the limited information these images contain. Thus, subsampled data from a Poisson resampling method may be a possible solution to conduct further studies regarding the classification of low-time SPECT-MPI.
dirty 0
eu_rights_str_mv openAccess
format masterThesis
fulltext.url.fl_str_mv https://repositorio.ipl.pt/bitstreams/6923ccae-7d9b-479c-bc76-05a2c38f64b8/download
id ripl_c4e17ee00a6f41c8e0cae4be30e492e9
identifier.url.fl_str_mv http://hdl.handle.net/10400.21/17628
instacron_str ipl
institution Instituto Politécnico de Lisboa
instname_str Instituto Politécnico de Lisboa
language eng
network_acronym_str ripl
network_name_str Repositório Científico do Instituto Politécnico de Lisboa
oai_identifier_str oai:repositorio.ipl.pt:10400.21/17628
organization_str_mv urn:organizationAcronym:ipl
person_str_mv Gonçalves, Gonçalo Cravo de Jesus
publishDate 2023
publisher.none.fl_str_mv Instituto Politécnico de Lisboa, Escola Superior de Tecnologia da Saúde de Lisboa
reponame_str Repositório Científico do Instituto Politécnico de Lisboa
repository_id_str urn:repositoryAcronym:ripl
service_str_mv urn:repositoryAcronym:ripl
spelling engInstituto Politécnico de Lisboa, Escola Superior de Tecnologia da Saúde de Lisboapt_PTABSTRACT - Myocardial perfusion imaging (MPI) by single-photon emission computed tomography (SPECT) plays a crucial role in the diagnosis of coronary artery disease. Moreover, the quantification of these images typically involves the extraction of quantitative parameters obtained from the rest stress perfusion. However, the acquisition systems have some limitations such as spatial blurring and low-count data, which may introduce bias in the classification. Additionally, these processes are time-consuming and user-dependent, leading to significant intra and inter-operator variability. Furthermore, over the years there has been a constant effort to reduce the dose of MPI. In this sense, both the variability classification issues and the dose reduction concerns can impact the true assessment of SPECT-MPI. In recent years, with the rise of artificial intelligence algorithms, several studies have proposed automatic Deep Learning techniques for the classification of MPI, moreover regarding low-count data. In this project, we ran 5 Convolutional Neural Network models with pre-trained weights: one trained on stress real full-time data (100%, as 100R), three individual models with synthetic 75%, 50%, and 25% count settings, and another one with all datasets combined (ALL). Thus, we compared their performance when tested on full-time and low-time studies and assessed the application of synthetic subsampled data from the Poisson Resampling technique in SPECT-MPI classification tasks. In conclusion, both 100R and ALL models achieved good and similar results when tested in real full-time (the 100R model achieved an accuracy of 0.70 and the ALL model achieved an accuracy of 0.65) and real low-time at 75% (both models achieved an accuracy of 0.71). Bellow this percentage, the models’ accuracy began to drop, possibly due to the limited information these images contain. Thus, subsampled data from a Poisson resampling method may be a possible solution to conduct further studies regarding the classification of low-time SPECT-MPI.application/pdfpt_PTConvolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose studyGonçalves, Gonçalo Cravo de JesusFigueiredo, SérgioJorge, Pedro Miguel Torres MendesHostingInstitutionOrganizationalRCIPLe-mailmailto:rcaap@sp.ipl.ptrcaap@sp.ipl.pt2024-08-28T09:29:27Z2023-122023-12-01T00:00:00ZHandlehttp://hdl.handle.net/10400.21/17628http://purl.org/coar/access_right/c_abf2open accessDeep learningConvolutional neural networksImage classification SPECTMyocardial perfusion imagingNuclear medicineAprendizagem profundaRedes neuronais convolucionaisClassificação de imagens SPECTCintigrafia de perfusão do miocárdioMedicina nuclearMEB5812197 bytesliteraturehttp://purl.org/coar/resource_type/c_bdccmaster thesis2023-12http://creativecommons.org/licenses/by-nc-nd/4.0/http://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://repositorio.ipl.pt/bitstreams/6923ccae-7d9b-479c-bc76-05a2c38f64b8/download
spellingShingle Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
Gonçalves, Gonçalo Cravo de Jesus
Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
subject.fl_str_mv Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
title Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
title_full Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
title_fullStr Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
title_full_unstemmed Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
title_short Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
title_sort Convolutional neural networks for myocardial perfusion SPECT imaging classification: a full and low-dose study
topic Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
topic_facet Deep learning
Convolutional neural networks
Image classification SPECT
Myocardial perfusion imaging
Nuclear medicine
Aprendizagem profunda
Redes neuronais convolucionais
Classificação de imagens SPECT
Cintigrafia de perfusão do miocárdio
Medicina nuclear
MEB
url http://hdl.handle.net/10400.21/17628
visible 1