Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling

Pathology diagnosis of colorectal cancer is time-consuming and requires a high level of expertise. However, it is an essential step towards establishing the adequate treatment. The need to analyse a large number of these histopathological images calls for automatic tools capable of aiding pathologis...

Descripción completa

Detalles Bibliográficos
Autores: Socolovsky, Miguel, López, Alberto, Greenson, Joel K., Rennert, Gad, Gruber, Stephen B., Moreno Aguado, Víctor
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2025
País:España
Institución:Universidad de Barcelona
Repositorio:Dipòsit Digital de la UB
OAI Identifier:oai:diposit.ub.edu:2445/224338
Acceso en línea:https://hdl.handle.net/2445/224338
Access Level:acceso abierto
Palabra clave:Histologia
Aprenentatge profund
Càncer colorectal
Histology
Deep learning (Machine learning)
Colorectal cancer
id ES_3f39decab4a65e6344ebfb4cda9fa7a6
oai_identifier_str oai:diposit.ub.edu:2445/224338
network_acronym_str ES
network_name_str España
repository_id_str
spelling Efficient deep neural networks for cancer detection on histopathology combining attention and image downsamplingSocolovsky, MiguelLópez, AlbertoGreenson, Joel K.Rennert, Gad Gruber, Stephen B.Moreno Aguado, VíctorHistologiaAprenentatge profundCàncer colorectalHistologyDeep learning (Machine learning)Colorectal cancerPathology diagnosis of colorectal cancer is time-consuming and requires a high level of expertise. However, it is an essential step towards establishing the adequate treatment. The need to analyse a large number of these histopathological images calls for automatic tools capable of aiding pathologists in this arduous task. Deep learning techniques, together with the wealth of data available nowadays, provide a promising candidate for such job. Adopting state-of-the-art artificial intelligence algorithms, we developed a model to accurately detect colorectal cancer in digitalised histopathological whole-slide images. Our end-to-end approach uses the principles of multiple-instance learning combined with deep convolutional neural networks in order to fully leverage the information contained within each image and make robust predictions at the patient's level. The model also allows to highlight the areas in the slide most likely to harbour tumour tissue. Given the finite computational resources available, working at maximum resolution can be detrimental. Therefore, we explored the impact of lowering the working image resolution. The algorithms were trained and validated on a subset of more than 1300 patients of the Molecular Epidemiology of Colorectal Cancer study with histopathology images available. These images gave rise to > 10(5) tiles of 256 x 256 pixels each. Once we identified the best-performing model we put it to the test on images from The Cancer Genome Atlas. We obtained the best outcomes working at 4 mu m/pix, achieving the following metrics on the test dataset: F1-Score of 0.96, a Matthews correlation coefficient of 0.92 and an area under the receiver operating characteristic curve of 0.99. These results are exceptional and prove that computational costs can be reduced while keeping the performance up to standard.Springer Science and Business Media LLC2025info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://hdl.handle.net/2445/224338Articles publicats en revistes (Institut d'lnvestigació Biomèdica de Bellvitge (IDIBELL))reponame:Dipòsit Digital de la UBinstname:Universidad de BarcelonaInglésReproducció del document publicat a: https://doi.org/10.1038/s41598-025-20954-2Scientific Reports, 2025, vol. 15, 36917https://doi.org/10.1038/s41598-025-20954-2cc-by (c) Socolovsky, Miguel et al., 2025https://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:diposit.ub.edu:2445/2243382026-05-27T06:46:51Z
dc.title.none.fl_str_mv Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
title Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
spellingShingle Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
Socolovsky, Miguel
Histologia
Aprenentatge profund
Càncer colorectal
Histology
Deep learning (Machine learning)
Colorectal cancer
title_short Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
title_full Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
title_fullStr Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
title_full_unstemmed Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
title_sort Efficient deep neural networks for cancer detection on histopathology combining attention and image downsampling
dc.creator.none.fl_str_mv Socolovsky, Miguel
López, Alberto
Greenson, Joel K.
Rennert, Gad
Gruber, Stephen B.
Moreno Aguado, Víctor
author Socolovsky, Miguel
author_facet Socolovsky, Miguel
López, Alberto
Greenson, Joel K.
Rennert, Gad
Gruber, Stephen B.
Moreno Aguado, Víctor
author_role author
author2 López, Alberto
Greenson, Joel K.
Rennert, Gad
Gruber, Stephen B.
Moreno Aguado, Víctor
author2_role author
author
author
author
author
dc.subject.none.fl_str_mv Histologia
Aprenentatge profund
Càncer colorectal
Histology
Deep learning (Machine learning)
Colorectal cancer
topic Histologia
Aprenentatge profund
Càncer colorectal
Histology
Deep learning (Machine learning)
Colorectal cancer
description Pathology diagnosis of colorectal cancer is time-consuming and requires a high level of expertise. However, it is an essential step towards establishing the adequate treatment. The need to analyse a large number of these histopathological images calls for automatic tools capable of aiding pathologists in this arduous task. Deep learning techniques, together with the wealth of data available nowadays, provide a promising candidate for such job. Adopting state-of-the-art artificial intelligence algorithms, we developed a model to accurately detect colorectal cancer in digitalised histopathological whole-slide images. Our end-to-end approach uses the principles of multiple-instance learning combined with deep convolutional neural networks in order to fully leverage the information contained within each image and make robust predictions at the patient's level. The model also allows to highlight the areas in the slide most likely to harbour tumour tissue. Given the finite computational resources available, working at maximum resolution can be detrimental. Therefore, we explored the impact of lowering the working image resolution. The algorithms were trained and validated on a subset of more than 1300 patients of the Molecular Epidemiology of Colorectal Cancer study with histopathology images available. These images gave rise to > 10(5) tiles of 256 x 256 pixels each. Once we identified the best-performing model we put it to the test on images from The Cancer Genome Atlas. We obtained the best outcomes working at 4 mu m/pix, achieving the following metrics on the test dataset: F1-Score of 0.96, a Matthews correlation coefficient of 0.92 and an area under the receiver operating characteristic curve of 0.99. These results are exceptional and prove that computational costs can be reduced while keeping the performance up to standard.
publishDate 2025
dc.date.none.fl_str_mv 2025
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://hdl.handle.net/2445/224338
url https://hdl.handle.net/2445/224338
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Reproducció del document publicat a: https://doi.org/10.1038/s41598-025-20954-2
Scientific Reports, 2025, vol. 15, 36917
https://doi.org/10.1038/s41598-025-20954-2
dc.rights.none.fl_str_mv cc-by (c) Socolovsky, Miguel et al., 2025
https://creativecommons.org/licenses/by/4.0/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv cc-by (c) Socolovsky, Miguel et al., 2025
https://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Springer Science and Business Media LLC
publisher.none.fl_str_mv Springer Science and Business Media LLC
dc.source.none.fl_str_mv Articles publicats en revistes (Institut d'lnvestigació Biomèdica de Bellvitge (IDIBELL))
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_ 1869406618894991360
score 15.81155