Abdominal MRI Unconditional Synthesis with Medical Assessment

Bibliographic Details
Main Author: Gonçalves, Bernardo
Publication Date: 2024
Other Authors: Silva, Mariana, Vieira, Luísa, Vieira, Pedro
Format: Article
Language: eng
Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
Download full: http://hdl.handle.net/10362/172413
Summary: Funding Information: This work was funded by the FCT\u2014Portuguese Foundation for Science and Technology and Bee2Fire SA under a PhD grant with reference PD/BDE/150624/2020. Publisher Copyright: © 2024 by the authors.
id RCAP_5e6c8e03aed7d3badec018b7612ed8a7
oai_identifier_str oai:run.unl.pt:10362/172413
network_acronym_str RCAP
network_name_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository_id_str https://opendoar.ac.uk/repository/7160
spelling Abdominal MRI Unconditional Synthesis with Medical Assessmentgenerative adversarial networksmedical imaging synthesisMRI synthesisStyleGAN3unconditional synthesisComputer Science (miscellaneous)Medicine (miscellaneous)Health InformaticsHealth Professions (miscellaneous)Funding Information: This work was funded by the FCT\u2014Portuguese Foundation for Science and Technology and Bee2Fire SA under a PhD grant with reference PD/BDE/150624/2020. Publisher Copyright: © 2024 by the authors.Current computer vision models require a significant amount of annotated data to improve their performance in a particular task. However, obtaining the required annotated data is challenging, especially in medicine. Hence, data augmentation techniques play a crucial role. In recent years, generative models have been used to create artificial medical images, which have shown promising results. This study aimed to use a state-of-the-art generative model, StyleGAN3, to generate realistic synthetic abdominal magnetic resonance images. These images will be evaluated using quantitative metrics and qualitative assessments by medical professionals. For this purpose, an abdominal MRI dataset acquired at Garcia da Horta Hospital in Almada, Portugal, was used. A subset containing only axial gadolinium-enhanced slices was used to train the model. The obtained Fréchet inception distance value (12.89) aligned with the state of the art, and a medical expert confirmed the significant realism and quality of the images. However, specific issues were identified in the generated images, such as texture variations, visual artefacts and anatomical inconsistencies. Despite these, this work demonstrated that StyleGAN3 is a viable solution to synthesise realistic medical imaging data, particularly in abdominal imaging.DF – Departamento de FísicaRUNGonçalves, BernardoSilva, MarianaVieira, LuísaVieira, Pedro2024-09-25T22:27:19Z2024-06-072024-06-07T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article13application/pdfhttp://hdl.handle.net/10362/172413eng2673-7426PURE: 99834574https://doi.org/10.3390/biomedinformatics4020082info:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2024-09-30T01:43:12Zoai:run.unl.pt:10362/172413Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-28T18:54:23.009860Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse
dc.title.none.fl_str_mv Abdominal MRI Unconditional Synthesis with Medical Assessment
title Abdominal MRI Unconditional Synthesis with Medical Assessment
spellingShingle Abdominal MRI Unconditional Synthesis with Medical Assessment
Gonçalves, Bernardo
generative adversarial networks
medical imaging synthesis
MRI synthesis
StyleGAN3
unconditional synthesis
Computer Science (miscellaneous)
Medicine (miscellaneous)
Health Informatics
Health Professions (miscellaneous)
title_short Abdominal MRI Unconditional Synthesis with Medical Assessment
title_full Abdominal MRI Unconditional Synthesis with Medical Assessment
title_fullStr Abdominal MRI Unconditional Synthesis with Medical Assessment
title_full_unstemmed Abdominal MRI Unconditional Synthesis with Medical Assessment
title_sort Abdominal MRI Unconditional Synthesis with Medical Assessment
author Gonçalves, Bernardo
author_facet Gonçalves, Bernardo
Silva, Mariana
Vieira, Luísa
Vieira, Pedro
author_role author
author2 Silva, Mariana
Vieira, Luísa
Vieira, Pedro
author2_role author
author
author
dc.contributor.none.fl_str_mv DF – Departamento de Física
RUN
dc.contributor.author.fl_str_mv Gonçalves, Bernardo
Silva, Mariana
Vieira, Luísa
Vieira, Pedro
dc.subject.por.fl_str_mv generative adversarial networks
medical imaging synthesis
MRI synthesis
StyleGAN3
unconditional synthesis
Computer Science (miscellaneous)
Medicine (miscellaneous)
Health Informatics
Health Professions (miscellaneous)
topic generative adversarial networks
medical imaging synthesis
MRI synthesis
StyleGAN3
unconditional synthesis
Computer Science (miscellaneous)
Medicine (miscellaneous)
Health Informatics
Health Professions (miscellaneous)
description Funding Information: This work was funded by the FCT\u2014Portuguese Foundation for Science and Technology and Bee2Fire SA under a PhD grant with reference PD/BDE/150624/2020. Publisher Copyright: © 2024 by the authors.
publishDate 2024
dc.date.none.fl_str_mv 2024-09-25T22:27:19Z
2024-06-07
2024-06-07T00:00:00Z
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://hdl.handle.net/10362/172413
url http://hdl.handle.net/10362/172413
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv 2673-7426
PURE: 99834574
https://doi.org/10.3390/biomedinformatics4020082
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 13
application/pdf
dc.source.none.fl_str_mv reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron:RCAAP
instname_str FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron_str RCAAP
institution RCAAP
reponame_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
collection Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository.name.fl_str_mv Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
repository.mail.fl_str_mv info@rcaap.pt
_version_ 1833597746707169280

Similar Items