A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , |
Tipo de documento: | Artigo de conferência |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1117/12.2611833 http://hdl.handle.net/11449/240247 |
Resumo: | Digital Breast Tomosynthesis (DBT) projections are acquired with a high level of noise, compared to Digital Mammography (DM) projections. Noise reduction in DBT projections is important because the projections are obtained with low radiation dose, elevating the noise level. In this way, noise reduction is essential to improve the quality of DBT exam. Recently, neural network based methods have been applied to denoise DBT projections, reaching remarkable results. Some papers have been published showing that these methods are able to overpass traditional methods’ results, but we could not find a paper comparing the different types of networks to denoise DBT projections. In this paper, we proposed an experiment to compare neural network based methods, with different architecture types, and traditional methods. We performed a comparison among five traditional non-blind denoising methods and six neural network models. Considering both quantitative and qualitative analysis, we found that some neural network models achieve remarkable results, especially shallower models. |
id |
UNSP_94affe45c0764a1c1d46e392ebebfd65 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/240247 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methodsconvolutional neural networksdeep learningDenoisingdigital breast tomosynthesisDigital Breast Tomosynthesis (DBT) projections are acquired with a high level of noise, compared to Digital Mammography (DM) projections. Noise reduction in DBT projections is important because the projections are obtained with low radiation dose, elevating the noise level. In this way, noise reduction is essential to improve the quality of DBT exam. Recently, neural network based methods have been applied to denoise DBT projections, reaching remarkable results. Some papers have been published showing that these methods are able to overpass traditional methods’ results, but we could not find a paper comparing the different types of networks to denoise DBT projections. In this paper, we proposed an experiment to compare neural network based methods, with different architecture types, and traditional methods. We performed a comparison among five traditional non-blind denoising methods and six neural network models. Considering both quantitative and qualitative analysis, we found that some neural network models achieve remarkable results, especially shallower models.São Paulo State University (Unesp) Institute of Geosciences and Exact Sciences (IGCE), SPSão Paulo State University (Unesp) Institute of Geosciences and Exact Sciences (IGCE), SPUniversidade Estadual Paulista (UNESP)Araújo, Darlan M.N. de [UNESP]Salvadeo, Denis H.P. [UNESP]Paula, Davi D. de [UNESP]2023-03-01T20:08:11Z2023-03-01T20:08:11Z2022-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObjecthttp://dx.doi.org/10.1117/12.2611833Progress in Biomedical Optics and Imaging - Proceedings of SPIE, v. 12032.1605-7422http://hdl.handle.net/11449/24024710.1117/12.26118332-s2.0-85131951516Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengProgress in Biomedical Optics and Imaging - Proceedings of SPIEinfo:eu-repo/semantics/openAccess2023-03-01T20:08:12Zoai:repositorio.unesp.br:11449/240247Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T14:39:18.212939Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
title |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
spellingShingle |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods Araújo, Darlan M.N. de [UNESP] convolutional neural networks deep learning Denoising digital breast tomosynthesis |
title_short |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
title_full |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
title_fullStr |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
title_full_unstemmed |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
title_sort |
A benchmark of denoising Digital Breast Tomosynthesis in projection domain: neural network-based and traditional methods |
author |
Araújo, Darlan M.N. de [UNESP] |
author_facet |
Araújo, Darlan M.N. de [UNESP] Salvadeo, Denis H.P. [UNESP] Paula, Davi D. de [UNESP] |
author_role |
author |
author2 |
Salvadeo, Denis H.P. [UNESP] Paula, Davi D. de [UNESP] |
author2_role |
author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (UNESP) |
dc.contributor.author.fl_str_mv |
Araújo, Darlan M.N. de [UNESP] Salvadeo, Denis H.P. [UNESP] Paula, Davi D. de [UNESP] |
dc.subject.por.fl_str_mv |
convolutional neural networks deep learning Denoising digital breast tomosynthesis |
topic |
convolutional neural networks deep learning Denoising digital breast tomosynthesis |
description |
Digital Breast Tomosynthesis (DBT) projections are acquired with a high level of noise, compared to Digital Mammography (DM) projections. Noise reduction in DBT projections is important because the projections are obtained with low radiation dose, elevating the noise level. In this way, noise reduction is essential to improve the quality of DBT exam. Recently, neural network based methods have been applied to denoise DBT projections, reaching remarkable results. Some papers have been published showing that these methods are able to overpass traditional methods’ results, but we could not find a paper comparing the different types of networks to denoise DBT projections. In this paper, we proposed an experiment to compare neural network based methods, with different architecture types, and traditional methods. We performed a comparison among five traditional non-blind denoising methods and six neural network models. Considering both quantitative and qualitative analysis, we found that some neural network models achieve remarkable results, especially shallower models. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-01-01 2023-03-01T20:08:11Z 2023-03-01T20:08:11Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/conferenceObject |
format |
conferenceObject |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1117/12.2611833 Progress in Biomedical Optics and Imaging - Proceedings of SPIE, v. 12032. 1605-7422 http://hdl.handle.net/11449/240247 10.1117/12.2611833 2-s2.0-85131951516 |
url |
http://dx.doi.org/10.1117/12.2611833 http://hdl.handle.net/11449/240247 |
identifier_str_mv |
Progress in Biomedical Optics and Imaging - Proceedings of SPIE, v. 12032. 1605-7422 10.1117/12.2611833 2-s2.0-85131951516 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Progress in Biomedical Optics and Imaging - Proceedings of SPIE |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808128394898964480 |