Exploring different Convolutional Neural Networks architectures to identify cells in spheroids
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Outros Autores: | , , , , , |
Tipo de documento: | Artigo de conferência |
Título da fonte: | Repositório Institucional do IPEN |
Texto Completo: | http://200.136.52.105/handle/123456789/31684 |
Resumo: | The cultivation of cells in 3D has gained more interest in research once 3D architecture can be closer to full cell physiological functionality. The cultivation of the cells in a spheroid format has shown very promising results, further for bioprinting developing so fast during the last decade. The interaction of spheroids and the matrix, or bioink, have proportionate new structures to be analyzed, specially if one would like to follow the whole system (spheroid and bioink) without fluorescent dyes. Trying to solve this image limitation, the aim of this paper is to present a study on different Convolutional Neural Networks (CNN) architectures employed to identify different structures in fibroblast NIH-3T3 spheroids. Three different architectures were considered: GoogleNet, ResNet18 and AlexNet, all implemented in Python 3.7 using the PyTorch Application Interface Programming (API). Given a spheroid image taken in a light microscope, four structures can be identified: the cell, the dead cell, the impurity/contamination and the background consisting of a gel in which the spheroid is immersed. All four CNN architectures were trained and evaluated with a dataset consisting of over 370 samples, split into a training set (??? 70%), a test set (??? 20%) and a validation set (??? 10%). Since our dataset has unbalanced classes, a data augmentation was applied in order to provide a comparable number of samples for all classes being considered. |
id |
IPEN_0222fdd850d6d97b50a08f8ba83625ef |
---|---|
oai_identifier_str |
oai:repositorio.ipen.br:123456789/31684 |
network_acronym_str |
IPEN |
network_name_str |
Repositório Institucional do IPEN |
repository_id_str |
4510 |
spelling |
2020-12-18T19:16:04Z2020-12-18T19:16:04ZOctober 26-30, 2020http://200.136.52.105/handle/123456789/31684The cultivation of cells in 3D has gained more interest in research once 3D architecture can be closer to full cell physiological functionality. The cultivation of the cells in a spheroid format has shown very promising results, further for bioprinting developing so fast during the last decade. The interaction of spheroids and the matrix, or bioink, have proportionate new structures to be analyzed, specially if one would like to follow the whole system (spheroid and bioink) without fluorescent dyes. Trying to solve this image limitation, the aim of this paper is to present a study on different Convolutional Neural Networks (CNN) architectures employed to identify different structures in fibroblast NIH-3T3 spheroids. Three different architectures were considered: GoogleNet, ResNet18 and AlexNet, all implemented in Python 3.7 using the PyTorch Application Interface Programming (API). Given a spheroid image taken in a light microscope, four structures can be identified: the cell, the dead cell, the impurity/contamination and the background consisting of a gel in which the spheroid is immersed. All four CNN architectures were trained and evaluated with a dataset consisting of over 370 samples, split into a training set (??? 70%), a test set (??? 20%) and a validation set (??? 10%). Since our dataset has unbalanced classes, a data augmentation was applied in order to provide a comparable number of samples for all classes being considered.Submitted by Pedro Silva Filho (pfsilva@ipen.br) on 2020-12-18T19:16:04Z No. of bitstreams: 1 27456.pdf: 647421 bytes, checksum: ed8cc8d736b846320ca284840df622e7 (MD5)Made available in DSpace on 2020-12-18T19:16:04Z (GMT). No. of bitstreams: 1 27456.pdf: 647421 bytes, checksum: ed8cc8d736b846320ca284840df622e7 (MD5)Coordena????o de Aperfei??oamento de Pessoal de N??vel Superior (CAPES)CAPES: 0012256-2260Sociedade Brasileira de Engenharia Biom??dicaanimal cellsspheroidsneural networkscell culturesfibroblastsimage processingExploring different Convolutional Neural Networks architectures to identify cells in spheroidsinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObjectCBEBIRio de Janeiro, RJVit??ria, ESSANTIAGO, A.G.CAMPOS, C.S.MACEDO, M.M.G.DAGUANO, J.K.M.B.DERNOWSEK, J.A.RODAS, A.C.D.BRAZILIAN CONGRESS IN BIOMEDICAL ENGINEERING, 27thinfo:eu-repo/semantics/openAccessreponame:Repositório Institucional do IPENinstname:Instituto de Pesquisas Energéticas e Nucleares (IPEN)instacron:IPEN274562020DERNOWSEK, J.A.20-12Proceedings15433DERNOWSEK, J.A.:15433:-1:NORIGINAL27456.pdf27456.pdfapplication/pdf647421http://repositorio.ipen.br/bitstream/123456789/31684/1/27456.pdfed8cc8d736b846320ca284840df622e7MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-81748http://repositorio.ipen.br/bitstream/123456789/31684/2/license.txt8a4605be74aa9ea9d79846c1fba20a33MD52123456789/316842022-06-27 16:05:20.414oai:repositorio.ipen.br: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Repositório InstitucionalPUBhttp://repositorio.ipen.br/oai/requestbibl@ipen.bropendoar:45102022-06-27T16:05:20Repositório Institucional do IPEN - Instituto de Pesquisas Energéticas e Nucleares (IPEN)false |
dc.title.pt_BR.fl_str_mv |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
title |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
spellingShingle |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids SANTIAGO, A.G. animal cells spheroids neural networks cell cultures fibroblasts image processing |
title_short |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
title_full |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
title_fullStr |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
title_full_unstemmed |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
title_sort |
Exploring different Convolutional Neural Networks architectures to identify cells in spheroids |
author |
SANTIAGO, A.G. |
author_facet |
SANTIAGO, A.G. CAMPOS, C.S. MACEDO, M.M.G. DAGUANO, J.K.M.B. DERNOWSEK, J.A. RODAS, A.C.D. BRAZILIAN CONGRESS IN BIOMEDICAL ENGINEERING, 27th |
author_role |
author |
author2 |
CAMPOS, C.S. MACEDO, M.M.G. DAGUANO, J.K.M.B. DERNOWSEK, J.A. RODAS, A.C.D. BRAZILIAN CONGRESS IN BIOMEDICAL ENGINEERING, 27th |
author2_role |
author author author author author author |
dc.contributor.author.fl_str_mv |
SANTIAGO, A.G. CAMPOS, C.S. MACEDO, M.M.G. DAGUANO, J.K.M.B. DERNOWSEK, J.A. RODAS, A.C.D. BRAZILIAN CONGRESS IN BIOMEDICAL ENGINEERING, 27th |
dc.subject.por.fl_str_mv |
animal cells spheroids neural networks cell cultures fibroblasts image processing |
topic |
animal cells spheroids neural networks cell cultures fibroblasts image processing |
description |
The cultivation of cells in 3D has gained more interest in research once 3D architecture can be closer to full cell physiological functionality. The cultivation of the cells in a spheroid format has shown very promising results, further for bioprinting developing so fast during the last decade. The interaction of spheroids and the matrix, or bioink, have proportionate new structures to be analyzed, specially if one would like to follow the whole system (spheroid and bioink) without fluorescent dyes. Trying to solve this image limitation, the aim of this paper is to present a study on different Convolutional Neural Networks (CNN) architectures employed to identify different structures in fibroblast NIH-3T3 spheroids. Three different architectures were considered: GoogleNet, ResNet18 and AlexNet, all implemented in Python 3.7 using the PyTorch Application Interface Programming (API). Given a spheroid image taken in a light microscope, four structures can be identified: the cell, the dead cell, the impurity/contamination and the background consisting of a gel in which the spheroid is immersed. All four CNN architectures were trained and evaluated with a dataset consisting of over 370 samples, split into a training set (??? 70%), a test set (??? 20%) and a validation set (??? 10%). Since our dataset has unbalanced classes, a data augmentation was applied in order to provide a comparable number of samples for all classes being considered. |
publishDate |
2020 |
dc.date.evento.pt_BR.fl_str_mv |
October 26-30, 2020 |
dc.date.accessioned.fl_str_mv |
2020-12-18T19:16:04Z |
dc.date.available.fl_str_mv |
2020-12-18T19:16:04Z |
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://200.136.52.105/handle/123456789/31684 |
url |
http://200.136.52.105/handle/123456789/31684 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
2256-2260 |
dc.coverage.pt_BR.fl_str_mv |
I |
dc.publisher.none.fl_str_mv |
Sociedade Brasileira de Engenharia Biom??dica |
publisher.none.fl_str_mv |
Sociedade Brasileira de Engenharia Biom??dica |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional do IPEN instname:Instituto de Pesquisas Energéticas e Nucleares (IPEN) instacron:IPEN |
instname_str |
Instituto de Pesquisas Energéticas e Nucleares (IPEN) |
instacron_str |
IPEN |
institution |
IPEN |
reponame_str |
Repositório Institucional do IPEN |
collection |
Repositório Institucional do IPEN |
bitstream.url.fl_str_mv |
http://repositorio.ipen.br/bitstream/123456789/31684/1/27456.pdf http://repositorio.ipen.br/bitstream/123456789/31684/2/license.txt |
bitstream.checksum.fl_str_mv |
ed8cc8d736b846320ca284840df622e7 8a4605be74aa9ea9d79846c1fba20a33 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional do IPEN - Instituto de Pesquisas Energéticas e Nucleares (IPEN) |
repository.mail.fl_str_mv |
bibl@ipen.br |
_version_ |
1767254252868075520 |