Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization

[EN] Background and Objective: Mitotic activity is a crucial biomarker for diagnosing and predicting outcomes for different types of cancers, particularly breast cancer. However, manual mitosis counting is challenging and time-consuming for pathologists, with moderate reproducibility due to biopsy s...

Descripción completa

Detalles Bibliográficos
Autores: Fernández-Martín, Claudio, Silva-Rodríguez, Julio, Kiraz, Umay, Janssen, Emiel A.M., Morales, Sandra|||0000-0003-0763-1545, Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
Tipo de recurso: artículo
Fecha de publicación:2024
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/205870
Acceso en línea:https://riunet.upv.es/handle/10251/205870
Access Level:acceso abierto
Palabra clave:Mitosis detection
MAI estimation
Weakly supervised learning
Hard samples distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
03.- Garantizar una vida saludable y promover el bienestar para todos y todas en todas las edades
id ES_2f8a39f04466e2818f2349b26e8b83f4
oai_identifier_str oai:riunet.upv.es:10251/205870
network_acronym_str ES
network_name_str España
repository_id_str
dc.title.none.fl_str_mv Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
title Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
spellingShingle Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
Fernández-Martín, Claudio
Mitosis detection
MAI estimation
Weakly supervised learning
Hard samples distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
03.- Garantizar una vida saludable y promover el bienestar para todos y todas en todas las edades
title_short Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
title_full Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
title_fullStr Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
title_full_unstemmed Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
title_sort Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization
dc.creator.none.fl_str_mv Fernández-Martín, Claudio
Silva-Rodríguez, Julio
Kiraz, Umay
Janssen, Emiel A.M.
Morales, Sandra|||0000-0003-0763-1545
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author Fernández-Martín, Claudio
author_facet Fernández-Martín, Claudio
Silva-Rodríguez, Julio
Kiraz, Umay
Janssen, Emiel A.M.
Morales, Sandra|||0000-0003-0763-1545
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author_role author
author2 Silva-Rodríguez, Julio
Kiraz, Umay
Janssen, Emiel A.M.
Morales, Sandra|||0000-0003-0763-1545
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author2_role author
author
author
author
author
dc.contributor.none.fl_str_mv Escuela Técnica Superior de Ingeniería de Telecomunicación
Departamento de Matemática Aplicada
Departamento de Comunicaciones
Escuela Técnica Superior de Ingeniería Aeroespacial y Diseño Industrial
Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano
GENERALITAT VALENCIANA
COMISION DE LAS COMUNIDADES EUROPEA
UNIVERSIDAD POLITECNICA DE VALENCIA
Universitat Politècnica de València
Repositorio Institucional de la Universitat Politècnica de València Riunet
dc.subject.none.fl_str_mv Mitosis detection
MAI estimation
Weakly supervised learning
Hard samples distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
03.- Garantizar una vida saludable y promover el bienestar para todos y todas en todas las edades
topic Mitosis detection
MAI estimation
Weakly supervised learning
Hard samples distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
03.- Garantizar una vida saludable y promover el bienestar para todos y todas en todas las edades
description [EN] Background and Objective: Mitotic activity is a crucial biomarker for diagnosing and predicting outcomes for different types of cancers, particularly breast cancer. However, manual mitosis counting is challenging and time-consuming for pathologists, with moderate reproducibility due to biopsy slide size, low mitotic cell density, and pattern heterogeneity. In recent years, deep learning methods based on convolutional neural networks (CNNs) have been proposed to address these limitations. Nonetheless, these methods have been hampered by the available data labels, which usually consist only of the centroids of mitosis, and by the incoming noise from annotated hard negatives. As a result, complex algorithms with multiple stages are often required to refine the labels at the pixel level and reduce the number of false positives. Methods: This article presents a novel weakly supervised approach for mitosis detection that utilizes only image-level labels on histological hematoxylin and eosin (H&E) images, avoiding the need for complex labeling scenarios. Also, an Uninformed Teacher-Student (UTS) pipeline is introduced to detect and distill hard samples by comparing weakly supervised localizations and the annotated centroids, using strong augmentations to enhance uncertainty. Additionally, an automatic proliferation score is proposed that mimicks the pathologist-annotated mitotic activity index (MAI). The proposed approach is evaluated on three publicly available datasets for mitosis detection on breast histology samples, and two datasets for mitotic activity counting in whole-slide images. Results: The proposed framework achieves competitive performance with relevant prior literature in all the datasets used for evaluation without explicitly using the mitosis location information during training. This approach challenges previous methods that rely on strong mitosis location information and multiple stages to refine false positives. Furthermore, the proposed pipeline for hard-sample distillation demonstrates promising dataset-specific improvements. Concretely, when the annotation has not been thoroughly refined by multiple pathologists, the UTS model offers improvements of up to in mitosis localization, thanks to the detection and distillation of uncertain cases. Concerning the mitosis counting task, the proposed automatic proliferation score shows a moderate positive correlation with the MAI annotated by pathologists at the biopsy level on two external datasets. Conclusions: The proposed Uninformed Teacher-Student pipeline leverages strong augmentations to distill uncertain samples and measure dissimilarities between predicted and annotated mitosis. Results demonstrate the feasibility of the weakly supervised approach and highlight its potential as an objective evaluation tool for tumor proliferation.
publishDate 2024
dc.date.none.fl_str_mv 2024
2024-03-01
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://riunet.upv.es/handle/10251/205870
url https://riunet.upv.es/handle/10251/205870
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.relation.none.fl_str_mv European Commission https://doi.org/10.13039/501100000780 H2020 860627 CLoud ARtificial Intelligence For pathologY
Universitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-10-20 Análisis de imágenes histológicas basado en técnicas de inteligencia artificial - WSIA
Universitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-06-23 inteligencia artificial para la pREdicción del suBtipo molEcular del Cáncer de mAma
Generalitat Valenciana https://doi.org/10.13039/501100003359 CIPROM%2F2022%2F20 COMPUTACION Y TRATAMIENTO DE LA SEÑAL PARA LA SOCIEDAD Y LA INDUSTRIA DIGITALES
Universitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-12-23
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
Reconocimiento - No comercial - Sin obra derivada (by-nc-nd)
http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
Reconocimiento - No comercial - Sin obra derivada (by-nc-nd)
http://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
instname:Universitat Politècnica de València (UPV)
instname_str Universitat Politècnica de València (UPV)
reponame_str RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
collection RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869405485027819520
spelling Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localizationFernández-Martín, ClaudioSilva-Rodríguez, JulioKiraz, UmayJanssen, Emiel A.M.Morales, Sandra|||0000-0003-0763-1545Naranjo Ornedo, Valeriana|||0000-0002-0181-3412Mitosis detectionMAI estimationWeakly supervised learningHard samples distillationTEORÍA DE LA SEÑAL Y COMUNICACIONES03.- Garantizar una vida saludable y promover el bienestar para todos y todas en todas las edades[EN] Background and Objective: Mitotic activity is a crucial biomarker for diagnosing and predicting outcomes for different types of cancers, particularly breast cancer. However, manual mitosis counting is challenging and time-consuming for pathologists, with moderate reproducibility due to biopsy slide size, low mitotic cell density, and pattern heterogeneity. In recent years, deep learning methods based on convolutional neural networks (CNNs) have been proposed to address these limitations. Nonetheless, these methods have been hampered by the available data labels, which usually consist only of the centroids of mitosis, and by the incoming noise from annotated hard negatives. As a result, complex algorithms with multiple stages are often required to refine the labels at the pixel level and reduce the number of false positives. Methods: This article presents a novel weakly supervised approach for mitosis detection that utilizes only image-level labels on histological hematoxylin and eosin (H&E) images, avoiding the need for complex labeling scenarios. Also, an Uninformed Teacher-Student (UTS) pipeline is introduced to detect and distill hard samples by comparing weakly supervised localizations and the annotated centroids, using strong augmentations to enhance uncertainty. Additionally, an automatic proliferation score is proposed that mimicks the pathologist-annotated mitotic activity index (MAI). The proposed approach is evaluated on three publicly available datasets for mitosis detection on breast histology samples, and two datasets for mitotic activity counting in whole-slide images. Results: The proposed framework achieves competitive performance with relevant prior literature in all the datasets used for evaluation without explicitly using the mitosis location information during training. This approach challenges previous methods that rely on strong mitosis location information and multiple stages to refine false positives. Furthermore, the proposed pipeline for hard-sample distillation demonstrates promising dataset-specific improvements. Concretely, when the annotation has not been thoroughly refined by multiple pathologists, the UTS model offers improvements of up to in mitosis localization, thanks to the detection and distillation of uncertain cases. Concerning the mitosis counting task, the proposed automatic proliferation score shows a moderate positive correlation with the MAI annotated by pathologists at the biopsy level on two external datasets. Conclusions: The proposed Uninformed Teacher-Student pipeline leverages strong augmentations to distill uncertain samples and measure dissimilarities between predicted and annotated mitosis. Results demonstrate the feasibility of the weakly supervised approach and highlight its potential as an objective evaluation tool for tumor proliferation.This work was funded by the Horizon 2020 European Union research and innovation programme under the Marie Sklodowska Curie grant agreement No 860627 (CLARIFY Project) . The work of Sandra Morales has been co-funded by the Universitat Politecnica de Valencia, Spain through the program PAID-10-20. The work of J. Silva-Rodriguez was carried out during his previous position at Universitat Politecnica de Valencia. This work was partially funded by Generalitat Valenciana through project CIPROM/2022/20 and with Ayuda a Primeros Proyectos de Investigacion (PAID-06-23) , Vicerrectorado de Investigacion of the Universitat Politecnica de Valencia. Funding for open access charge: Universitat Politecnica de Valencia (PAID-12-23).ElsevierEscuela Técnica Superior de Ingeniería de TelecomunicaciónDepartamento de Matemática AplicadaDepartamento de ComunicacionesEscuela Técnica Superior de Ingeniería Aeroespacial y Diseño IndustrialInstituto Universitario de Investigación en Tecnología Centrada en el Ser HumanoGENERALITAT VALENCIANACOMISION DE LAS COMUNIDADES EUROPEAUNIVERSIDAD POLITECNICA DE VALENCIAUniversitat Politècnica de ValènciaRepositorio Institucional de la Universitat Politècnica de València Riunet20242024-03-01journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://riunet.upv.es/handle/10251/205870reponame:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valénciainstname:Universitat Politècnica de València (UPV)InglésengEuropean Commission https://doi.org/10.13039/501100000780 H2020 860627 CLoud ARtificial Intelligence For pathologYUniversitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-10-20 Análisis de imágenes histológicas basado en técnicas de inteligencia artificial - WSIAUniversitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-06-23 inteligencia artificial para la pREdicción del suBtipo molEcular del Cáncer de mAmaGeneralitat Valenciana https://doi.org/10.13039/501100003359 CIPROM%2F2022%2F20 COMPUTACION Y TRATAMIENTO DE LA SEÑAL PARA LA SOCIEDAD Y LA INDUSTRIA DIGITALESUniversitat Politècnica de València https://doi.org/10.13039/501100004233 PAID-12-23open accesshttp://purl.org/coar/access_right/c_abf2Reconocimiento - No comercial - Sin obra derivada (by-nc-nd) http://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessoai:riunet.upv.es:10251/2058702026-06-13T07:49:27Z
score 15.812455

Similares en LA Referencia