Publicação

Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images

Ver documento

Detalhes bibliográficos
Resumo:Scoliosis, characterized by a three-dimensional deformation of the spine with a curvature exceeding 10° in the coronal plane, presents a diagnostic challenge. Conventionally, the gold standard for scoliosis diagnosis and spinal curvature measurement relies on X-ray imaging and the calculation of Cobb Angles. However, the ionizing radiation associated with X-rays poses potential risks, including the development of cancer. Moreover, the frequent use of X-rays for monitoring scoliosis progression or treatment outcomes increases the cumulative risk of these adverse effects. In contrast, ultrasound imaging emerges as a promising alternative. It is devoid of ionizing radiation, cost-effective, and highly portable, offering the potential for increased scoliosis screening and detection. Nonetheless, ultrasound spine imaging encounters challenges, notably low contrast and speckled noise, which compromise image quality. In this research endeavour, the application of cutting-edge deep learning techniques to address the challenges present in ultrasound images was studied. A novel deep learning model, known as the Hybrid Attention Recurrent Residual U-Net (Hybrid R2AU-Net), is proposed. It is specifically designed for segmenting spine structures in ultrasound spine images. The Hybrid R2AU-Net is a five-layer deep learning model that includes dense connections, Recurrent Residual Blocks (RecRes Blocks), and attention mechanisms in its architecture. The outcomes of this study were highly encouraging. The Hybrid R2AU-Net outperformed alternative deep learning models, delivering superior results with a Dice Score of 85.2%, a Jaccard Index of 74.3%, and a Detection Rate of 92.4%. In the future, these remarkable achievements underscore the potential of the Hybrid R2AU-Net to be seamlessly integrated into an automated scoliosis diagnosis system, promising a more radiation-free and efficient approach to scoliosis management.
Autores principais:Monteiro, Maria Leonor Cohen
Assunto:Scoliosis Ultrasound Segmentation Deep Learning Convolutional Neural Network
Ano:2023
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL
_version_ 1868414792866201600
author Monteiro, Maria Leonor Cohen
author_facet Monteiro, Maria Leonor Cohen
author_role author
contributor_name_str_mv Ling, Steven
Vigário, Ricardo
RUN
country_str PT
creators_json_txt [{\"Person.name\":\"Monteiro, Maria Leonor Cohen\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Ling, Steven
Vigário, Ricardo
RUN
datacite.creators.creator.creatorName.fl_str_mv Monteiro, Maria Leonor Cohen
datacite.date.Accepted.fl_str_mv 2023-12-01T00:00:00Z
datacite.date.available.fl_str_mv 2024-02-16T11:33:35Z
datacite.date.embargoed.fl_str_mv 2024-02-16T11:33:35Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
datacite.titles.title.fl_str_mv Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
dc.contributor.none.fl_str_mv Ling, Steven
Vigário, Ricardo
RUN
dc.creator.none.fl_str_mv Monteiro, Maria Leonor Cohen
dc.date.Accepted.fl_str_mv 2023-12-01T00:00:00Z
dc.date.available.fl_str_mv 2024-02-16T11:33:35Z
dc.date.embargoed.fl_str_mv 2024-02-16T11:33:35Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10362/163649
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 Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
dc.title.fl_str_mv Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_bdcc
description Scoliosis, characterized by a three-dimensional deformation of the spine with a curvature exceeding 10° in the coronal plane, presents a diagnostic challenge. Conventionally, the gold standard for scoliosis diagnosis and spinal curvature measurement relies on X-ray imaging and the calculation of Cobb Angles. However, the ionizing radiation associated with X-rays poses potential risks, including the development of cancer. Moreover, the frequent use of X-rays for monitoring scoliosis progression or treatment outcomes increases the cumulative risk of these adverse effects. In contrast, ultrasound imaging emerges as a promising alternative. It is devoid of ionizing radiation, cost-effective, and highly portable, offering the potential for increased scoliosis screening and detection. Nonetheless, ultrasound spine imaging encounters challenges, notably low contrast and speckled noise, which compromise image quality. In this research endeavour, the application of cutting-edge deep learning techniques to address the challenges present in ultrasound images was studied. A novel deep learning model, known as the Hybrid Attention Recurrent Residual U-Net (Hybrid R2AU-Net), is proposed. It is specifically designed for segmenting spine structures in ultrasound spine images. The Hybrid R2AU-Net is a five-layer deep learning model that includes dense connections, Recurrent Residual Blocks (RecRes Blocks), and attention mechanisms in its architecture. The outcomes of this study were highly encouraging. The Hybrid R2AU-Net outperformed alternative deep learning models, delivering superior results with a Dice Score of 85.2%, a Jaccard Index of 74.3%, and a Detection Rate of 92.4%. In the future, these remarkable achievements underscore the potential of the Hybrid R2AU-Net to be seamlessly integrated into an automated scoliosis diagnosis system, promising a more radiation-free and efficient approach to scoliosis management.
dirty 0
eu_rights_str_mv openAccess
format masterThesis
fulltext.url.fl_str_mv https://run.unl.pt/bitstreams/f21e0a13-edd6-4f5b-b551-431afd4567dd/download
id run_ea4611ddd3dde19584f185ee3d3d5d7f
identifier.url.fl_str_mv http://hdl.handle.net/10362/163649
instacron_str unl
institution Universidade Nova de Lisboa
instname_str Universidade Nova de Lisboa
language eng
network_acronym_str run
network_name_str Repositório Institucional da UNL
oai_identifier_str oai:run.unl.pt:10362/163649
organization_str_mv urn:organizationAcronym:unl
person_str_mv Monteiro, Maria Leonor Cohen
publishDate 2023
reponame_str Repositório Institucional da UNL
repository_id_str urn:repositoryAcronym:run
service_str_mv urn:repositoryAcronym:run
spelling engpt_PTScoliosis, characterized by a three-dimensional deformation of the spine with a curvature exceeding 10° in the coronal plane, presents a diagnostic challenge. Conventionally, the gold standard for scoliosis diagnosis and spinal curvature measurement relies on X-ray imaging and the calculation of Cobb Angles. However, the ionizing radiation associated with X-rays poses potential risks, including the development of cancer. Moreover, the frequent use of X-rays for monitoring scoliosis progression or treatment outcomes increases the cumulative risk of these adverse effects. In contrast, ultrasound imaging emerges as a promising alternative. It is devoid of ionizing radiation, cost-effective, and highly portable, offering the potential for increased scoliosis screening and detection. Nonetheless, ultrasound spine imaging encounters challenges, notably low contrast and speckled noise, which compromise image quality. In this research endeavour, the application of cutting-edge deep learning techniques to address the challenges present in ultrasound images was studied. A novel deep learning model, known as the Hybrid Attention Recurrent Residual U-Net (Hybrid R2AU-Net), is proposed. It is specifically designed for segmenting spine structures in ultrasound spine images. The Hybrid R2AU-Net is a five-layer deep learning model that includes dense connections, Recurrent Residual Blocks (RecRes Blocks), and attention mechanisms in its architecture. The outcomes of this study were highly encouraging. The Hybrid R2AU-Net outperformed alternative deep learning models, delivering superior results with a Dice Score of 85.2%, a Jaccard Index of 74.3%, and a Detection Rate of 92.4%. In the future, these remarkable achievements underscore the potential of the Hybrid R2AU-Net to be seamlessly integrated into an automated scoliosis diagnosis system, promising a more radiation-free and efficient approach to scoliosis management.application/pdfpt_PTDeep Learning Approach for the Segmentation of Spinal Structures in Ultrasound ImagesMonteiro, Maria Leonor CohenLing, StevenVigário, RicardoHostingInstitutionOrganizationalRUNe-mailmailto:run@unl.ptrun@unl.pt2024-02-16T11:33:35Z2023-122023-12-01T00:00:00ZHandlehttp://hdl.handle.net/10362/163649http://purl.org/coar/access_right/c_abf2open accessScoliosisUltrasoundSegmentationDeep LearningConvolutional Neural Network4587210 bytesliteraturehttp://purl.org/coar/resource_type/c_bdccmaster thesishttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://run.unl.pt/bitstreams/f21e0a13-edd6-4f5b-b551-431afd4567dd/download
spellingShingle Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
Monteiro, Maria Leonor Cohen
Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
status SINGLETON
subject.fl_str_mv Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
title Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
title_full Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
title_fullStr Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
title_full_unstemmed Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
title_short Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
title_sort Deep Learning Approach for the Segmentation of Spinal Structures in Ultrasound Images
topic Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
topic_facet Scoliosis
Ultrasound
Segmentation
Deep Learning
Convolutional Neural Network
url http://hdl.handle.net/10362/163649
visible 1