Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach

Breast density assessed from digital mammograms is a known biomarker related to a higher risk of developing breast cancer. Supervised learning algorithms have been implemented to determine this. However, the performance of these algorithms depends on the quality of the ground-truth information, whic...

ver descrição completa

Detalhes bibliográficos
Autores: Larroza A, Pérez-Benito FJ, Perez-Cortes JC, Román M, Pollán M, Pérez-Gómez B, Salas-Trejo D, Llobet R
Formato: artículo
Estado:Versión publicada
Fecha de publicación:2022
País:España
Recursos:Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana (FISABIO)
Repositorio:r-FISABIO. Repositorio Institucional de Producción Científica
OAI Identifier:oai:fisabio.fundanetsuite.com:p13840
Acesso em linha:https://fisabio.portalinvestigacion.com/publicaciones/13840
Access Level:acceso abierto
Palavra-chave:mammography
breast density segmentation
deep learning
noisy labels
id ES_ed4f2ff613b7912f8fde2972ade1b5f7
oai_identifier_str oai:fisabio.fundanetsuite.com:p13840
network_acronym_str ES
network_name_str España
repository_id_str
spelling Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based ApproachLarroza APérez-Benito FJPerez-Cortes JCRomán MPollán MPérez-Gómez BSalas-Trejo DLlobet Rmammographybreast density segmentationdeep learningnoisy labelsBreast density assessed from digital mammograms is a known biomarker related to a higher risk of developing breast cancer. Supervised learning algorithms have been implemented to determine this. However, the performance of these algorithms depends on the quality of the ground-truth information, which expert readers usually provide. These expert labels are noisy approximations to the ground truth, as there is both intra- and inter-observer variability among them. Thus, it is crucial to provide a reliable method to measure breast density from mammograms. This paper presents a fully automated method based on deep learning to estimate breast density, including breast detection, pectoral muscle exclusion, and dense tissue segmentation. We propose a novel confusion matrix (CM)-YNet model for the segmentation step. This architecture includes networks to model each radiologist's noisy label and gives the estimated ground-truth segmentation as well as two parameters that allow interaction with a threshold-based labeling tool. A multi-center study involving 1785 women whose "for presentation" mammograms were obtained from 11 different medical facilities was performed. A total of 2496 mammograms were used as the training corpus, and 844 formed the testing corpus. Additionally, we included a totally independent dataset from a different center, composed of 381 women with one image per patient. Each mammogram was labeled independently by two expert radiologists using a threshold-based tool. The implemented CM-Ynet model achieved the highest DICE score averaged over both test datasets (0.82 +/- 0.14) when compared to the closest dense-tissue segmentation assessment from both radiologists. The level of concordance between the two radiologists showed a DICE score of 0.76 +/- 0.17. An automatic breast density estimator based on deep learning exhibited higher performance when compared with two experienced radiologists. This suggests that modeling each radiologist's label allows for better estimation of the unknown ground-truth segmentation. The advantage of the proposed model is that it also provides the threshold parameters that enable user interaction with a threshold-based tool.MDPI2022info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttps://fisabio.portalinvestigacion.com/publicaciones/13840DiagnosticsISSN: 20754418reponame:r-FISABIO. Repositorio Institucional de Producción Científicainstname:Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana (FISABIO)Inglésinfo:eu-repo/semantics/openAccessoai:fisabio.fundanetsuite.com:p138402026-06-11T12:45:17Z
dc.title.none.fl_str_mv Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
title Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
spellingShingle Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
Larroza A
mammography
breast density segmentation
deep learning
noisy labels
title_short Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
title_full Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
title_fullStr Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
title_full_unstemmed Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
title_sort Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
dc.creator.none.fl_str_mv Larroza A
Pérez-Benito FJ
Perez-Cortes JC
Román M
Pollán M
Pérez-Gómez B
Salas-Trejo D
Llobet R
author Larroza A
author_facet Larroza A
Pérez-Benito FJ
Perez-Cortes JC
Román M
Pollán M
Pérez-Gómez B
Salas-Trejo D
Llobet R
author_role author
author2 Pérez-Benito FJ
Perez-Cortes JC
Román M
Pollán M
Pérez-Gómez B
Salas-Trejo D
Llobet R
author2_role author
author
author
author
author
author
author
dc.subject.none.fl_str_mv mammography
breast density segmentation
deep learning
noisy labels
topic mammography
breast density segmentation
deep learning
noisy labels
description Breast density assessed from digital mammograms is a known biomarker related to a higher risk of developing breast cancer. Supervised learning algorithms have been implemented to determine this. However, the performance of these algorithms depends on the quality of the ground-truth information, which expert readers usually provide. These expert labels are noisy approximations to the ground truth, as there is both intra- and inter-observer variability among them. Thus, it is crucial to provide a reliable method to measure breast density from mammograms. This paper presents a fully automated method based on deep learning to estimate breast density, including breast detection, pectoral muscle exclusion, and dense tissue segmentation. We propose a novel confusion matrix (CM)-YNet model for the segmentation step. This architecture includes networks to model each radiologist's noisy label and gives the estimated ground-truth segmentation as well as two parameters that allow interaction with a threshold-based labeling tool. A multi-center study involving 1785 women whose "for presentation" mammograms were obtained from 11 different medical facilities was performed. A total of 2496 mammograms were used as the training corpus, and 844 formed the testing corpus. Additionally, we included a totally independent dataset from a different center, composed of 381 women with one image per patient. Each mammogram was labeled independently by two expert radiologists using a threshold-based tool. The implemented CM-Ynet model achieved the highest DICE score averaged over both test datasets (0.82 +/- 0.14) when compared to the closest dense-tissue segmentation assessment from both radiologists. The level of concordance between the two radiologists showed a DICE score of 0.76 +/- 0.17. An automatic breast density estimator based on deep learning exhibited higher performance when compared with two experienced radiologists. This suggests that modeling each radiologist's label allows for better estimation of the unknown ground-truth segmentation. The advantage of the proposed model is that it also provides the threshold parameters that enable user interaction with a threshold-based tool.
publishDate 2022
dc.date.none.fl_str_mv 2022
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://fisabio.portalinvestigacion.com/publicaciones/13840
url https://fisabio.portalinvestigacion.com/publicaciones/13840
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv MDPI
publisher.none.fl_str_mv MDPI
dc.source.none.fl_str_mv Diagnostics
ISSN: 20754418
reponame:r-FISABIO. Repositorio Institucional de Producción Científica
instname:Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana (FISABIO)
instname_str Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana (FISABIO)
reponame_str r-FISABIO. Repositorio Institucional de Producción Científica
collection r-FISABIO. Repositorio Institucional de Producción Científica
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869423443070418944
score 15,811543

Registos relacionados