Generalizability in multi-centre cardiac image analysis with machine learning

[eng] The field of Artificial Intelligence (AI) has undergone a revolution in recent years with the advent of more efficient computing hardware and well-documented software for model development. Many fields are being transformed. Medicine is one of the fields that has seen the appearance of models...

ver descrição completa

Detalhes bibliográficos
Autor: Campello Román, Víctor Manuel
Tipo de documento: tese
Estado:Versão publicada
Data de publicação:2023
País:España
Recursos:Universidad de Barcelona
Repositório:Dipòsit Digital de la UB
OAI Identifier:oai:diposit.ub.edu:2445/206020
Acesso em linha:https://hdl.handle.net/2445/206020
http://hdl.handle.net/10803/689810
Access Level:Acceso aberto
Palavra-chave:Aprenentatge automàtic
Ecocardiografia
Imatges per ressonància magnètica
Machine learning
Echocardiography
Magnetic resonance imaging
id ES_618a2cd6899c835be76fb8f5a16f1f7f
oai_identifier_str oai:diposit.ub.edu:2445/206020
network_acronym_str ES
network_name_str España
repository_id_str
spelling Generalizability in multi-centre cardiac image analysis with machine learningCampello Román, Víctor ManuelAprenentatge automàticEcocardiografiaImatges per ressonància magnèticaMachine learningEchocardiographyMagnetic resonance imaging[eng] The field of Artificial Intelligence (AI) has undergone a revolution in recent years with the advent of more efficient computing hardware and well-documented software for model development. Many fields are being transformed. Medicine is one of the fields that has seen the appearance of models that can solve complex tasks such as automatic image segmentation or diagnosis. However, there are important challenges that need to be overcome for a successful application in clinical practice. One important challenge is the generalization of models to unseen domains independently of other factors, such as the scanner manufacturer, the scanning protocol, the sample size or the image quality. In this thesis, we aim to investigate the effects of the domain shift in medical imaging, specifically for cardiovascular studies, which present a particular challenge since the heart is a moving organ. Furthermore, we aim to contribute to methods to overcome or reduce the model performance gap. First, we establish a collaboration with clinical researchers from six different centres from three countries and assemble a large multi-centre dataset to tackle one of the greatest challenges in research: the domain gap problem. We process and annotate the data and develop a benchmark study by organizing an international competition to compare and analyse different techniques to bridge the generalization gap. The dataset is later open-sourced to foster innovation within the research community, becoming the first open multi-centre cardiac dataset. Then, we perform an exhaustive comparison of domain generalization and adaptation methods, including the best-performing methods in the aforementioned competition, for late gadolinium- enhanced image segmentation for the first time. We show that extensive data augmentation is very important for generalization and that model fine-tuning can reach or even surpass multi-centre models. Finally, we investigate the effects of differences in image appearance for the first time in a multi-centre study with cardiovascular imaging and compare several harmonisation techniques both at the feature and image levels for improved diagnosis. We show that histogram matching-based harmonisation results in image features (radiomics) that are more generalizable across centres.Universitat de BarcelonaLekadir, Karim, 1977-Seguí Mesquida, SantiUniversitat de Barcelona. Facultat de Matemàtiques2023info:eu-repo/semantics/doctoralThesisinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://hdl.handle.net/2445/206020http://hdl.handle.net/10803/689810Tesis Doctorals - Facultat - Matemàtiquesreponame:Dipòsit Digital de la UBinstname:Universidad de BarcelonaIngléscc by-nc-nd (c) Campello Román, Víctor Manuel, 2024http://creativecommons.org/licenses/by-nc-nd/3.0/es/info:eu-repo/semantics/openAccessoai:diposit.ub.edu:2445/2060202026-05-27T06:46:51Z
dc.title.none.fl_str_mv Generalizability in multi-centre cardiac image analysis with machine learning
title Generalizability in multi-centre cardiac image analysis with machine learning
spellingShingle Generalizability in multi-centre cardiac image analysis with machine learning
Campello Román, Víctor Manuel
Aprenentatge automàtic
Ecocardiografia
Imatges per ressonància magnètica
Machine learning
Echocardiography
Magnetic resonance imaging
title_short Generalizability in multi-centre cardiac image analysis with machine learning
title_full Generalizability in multi-centre cardiac image analysis with machine learning
title_fullStr Generalizability in multi-centre cardiac image analysis with machine learning
title_full_unstemmed Generalizability in multi-centre cardiac image analysis with machine learning
title_sort Generalizability in multi-centre cardiac image analysis with machine learning
dc.creator.none.fl_str_mv Campello Román, Víctor Manuel
author Campello Román, Víctor Manuel
author_facet Campello Román, Víctor Manuel
author_role author
dc.contributor.none.fl_str_mv Lekadir, Karim, 1977-
Seguí Mesquida, Santi
Universitat de Barcelona. Facultat de Matemàtiques
dc.subject.none.fl_str_mv Aprenentatge automàtic
Ecocardiografia
Imatges per ressonància magnètica
Machine learning
Echocardiography
Magnetic resonance imaging
topic Aprenentatge automàtic
Ecocardiografia
Imatges per ressonància magnètica
Machine learning
Echocardiography
Magnetic resonance imaging
description [eng] The field of Artificial Intelligence (AI) has undergone a revolution in recent years with the advent of more efficient computing hardware and well-documented software for model development. Many fields are being transformed. Medicine is one of the fields that has seen the appearance of models that can solve complex tasks such as automatic image segmentation or diagnosis. However, there are important challenges that need to be overcome for a successful application in clinical practice. One important challenge is the generalization of models to unseen domains independently of other factors, such as the scanner manufacturer, the scanning protocol, the sample size or the image quality. In this thesis, we aim to investigate the effects of the domain shift in medical imaging, specifically for cardiovascular studies, which present a particular challenge since the heart is a moving organ. Furthermore, we aim to contribute to methods to overcome or reduce the model performance gap. First, we establish a collaboration with clinical researchers from six different centres from three countries and assemble a large multi-centre dataset to tackle one of the greatest challenges in research: the domain gap problem. We process and annotate the data and develop a benchmark study by organizing an international competition to compare and analyse different techniques to bridge the generalization gap. The dataset is later open-sourced to foster innovation within the research community, becoming the first open multi-centre cardiac dataset. Then, we perform an exhaustive comparison of domain generalization and adaptation methods, including the best-performing methods in the aforementioned competition, for late gadolinium- enhanced image segmentation for the first time. We show that extensive data augmentation is very important for generalization and that model fine-tuning can reach or even surpass multi-centre models. Finally, we investigate the effects of differences in image appearance for the first time in a multi-centre study with cardiovascular imaging and compare several harmonisation techniques both at the feature and image levels for improved diagnosis. We show that histogram matching-based harmonisation results in image features (radiomics) that are more generalizable across centres.
publishDate 2023
dc.date.none.fl_str_mv 2023
dc.type.none.fl_str_mv info:eu-repo/semantics/doctoralThesis
info:eu-repo/semantics/publishedVersion
format doctoralThesis
status_str publishedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/2445/206020
http://hdl.handle.net/10803/689810
url https://hdl.handle.net/2445/206020
http://hdl.handle.net/10803/689810
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.rights.none.fl_str_mv cc by-nc-nd (c) Campello Román, Víctor Manuel, 2024
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv cc by-nc-nd (c) Campello Román, Víctor Manuel, 2024
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universitat de Barcelona
publisher.none.fl_str_mv Universitat de Barcelona
dc.source.none.fl_str_mv Tesis Doctorals - Facultat - Matemàtiques
reponame:Dipòsit Digital de la UB
instname:Universidad de Barcelona
instname_str Universidad de Barcelona
reponame_str Dipòsit Digital de la UB
collection Dipòsit Digital de la UB
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869409423984689152
score 15,300724