MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images

[EN] Artificial intelligence (AI) agents encounter the problem of catastrophic forgetting when they are trained in sequentially with new data batches. This issue poses a barrier to the implementation of AI -based models in tasks that involve ongoing evolution, such as cancer prediction. Moreover, wh...

Descripción completa

Detalles Bibliográficos
Autores: Meseguer-Esbrí, Pablo|||0000-0001-7821-6168, del Amor, Rocío, 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:10251/214706
Acceso en línea:https://riunet.upv.es/handle/10251/214706
Access Level:acceso abierto
Palabra clave:Skin cancer
Whole slide images
Multiple instance learning
Class-incremental learning
Knowledge distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
id ES_a3f98b2425a973ba17c3cbf1e4ffb49a
oai_identifier_str oai:riunet.upv.es:10251/214706
network_acronym_str ES
network_name_str España
repository_id_str
dc.title.none.fl_str_mv MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
title MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
spellingShingle MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
Skin cancer
Whole slide images
Multiple instance learning
Class-incremental learning
Knowledge distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
title_short MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
title_full MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
title_fullStr MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
title_full_unstemmed MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
title_sort MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide images
dc.creator.none.fl_str_mv Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
del Amor, Rocío
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
author_facet Meseguer-Esbrí, Pablo|||0000-0001-7821-6168
del Amor, Rocío
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author_role author
author2 del Amor, Rocío
Naranjo Ornedo, Valeriana|||0000-0002-0181-3412
author2_role author
author
dc.contributor.none.fl_str_mv Escuela Técnica Superior de Ingeniería de Telecomunicación
Departamento de Comunicaciones
Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano
Escuela Técnica Superior de Ingeniería Industrial
Agencia Estatal de Investigación
• Agència Valenciana de la Innovació
European Commission
Universitat Politècnica de València
MINISTERIO DE UNIVERSIDADES E INVESTIGACION
Generalitat Valenciana
Repositorio Institucional de la Universitat Politècnica de València Riunet
dc.subject.none.fl_str_mv Skin cancer
Whole slide images
Multiple instance learning
Class-incremental learning
Knowledge distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
topic Skin cancer
Whole slide images
Multiple instance learning
Class-incremental learning
Knowledge distillation
TEORÍA DE LA SEÑAL Y COMUNICACIONES
description [EN] Artificial intelligence (AI) agents encounter the problem of catastrophic forgetting when they are trained in sequentially with new data batches. This issue poses a barrier to the implementation of AI -based models in tasks that involve ongoing evolution, such as cancer prediction. Moreover, whole slide images (WSI) play a crucial role in cancer management, and their automated analysis has become increasingly popular in assisting pathologists during the diagnosis process. Incremental learning (IL) techniques aim to develop algorithms capable of retaining previously acquired information while also acquiring new insights to predict future data. Deep IL techniques need to address the challenges posed by the gigapixel scale of WSIs, which often necessitates the use of multiple instance learning (MIL) frameworks. In this paper, we introduce an IL algorithm tailored for analyzing WSIs within a MIL paradigm. The proposed M ultiple I nstance C lass- I ncremental L earning ( MICIL ) algorithm combines MIL with class-IL for the first time, allowing for the incremental prediction of multiple skin cancer subtypes from WSIs within a class-IL scenario. Our framework incorporates knowledge distillation and data rehearsal, along with a novel embedding-level distillation, aiming to preserve the latent space at the aggregated WSI level. Results demonstrate the algorithm's effectiveness in addressing the challenge of balancing IL-specific metrics, such as intransigence and forgetting, and solving the plasticity-stability dilemma.
publishDate 2024
dc.date.none.fl_str_mv 2024
2024-06-01
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/10251/214706
url https://riunet.upv.es/handle/10251/214706
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.relation.none.fl_str_mv Agencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020 PID2019-105142RB-C21 CARACTERIZACION DE NEOPLASIAS DE CELULAS FUSIFORMES EN IMAGENES HISTOLOGICAS
Agencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 PID2022-140189OB-C21 RECUPERACION DE IMAGENES BASADA EN EL CONTENIDO PARA EL DIAGNOSTICO DE TUMORES CUTANEOS PRIMARIOS Y SECUNDARIOS
European Commission https://doi.org/10.13039/501100000780 H2020 860627 CLoud ARtificial Intelligence For pathologY
Agència Valenciana de la Innovació https://doi.org/10.13039/501100016028 INNEST%2F2021%2F321 SISTEMA DE VISIÓN ARTIFICIAL AUMENTADA PARA LA CARACTERIZACIÓN MOLECULAR Y MORFOLÓGICA DEL CÁNCER DE PIEL
Ministerio de Ciencia, Innovación y Universidades https://doi.org/10.13039/100014440 FPU20%2F05263 Sistema automático de clasificación de neoplasias cutáneas de células fusiformes basado en inteligencia artificial
Generalitat Valenciana https://doi.org/10.13039/501100003359 VALGRAI%2F22%2F4 Estrategias de aprendizaje profundo para el análisis de imágenes histopatológicas
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_ 1869415441563123712
spelling MICIL: Multiple-Instance Class-Incremental Learning for skin cancer whole slide imagesMeseguer-Esbrí, Pablo|||0000-0001-7821-6168del Amor, RocíoNaranjo Ornedo, Valeriana|||0000-0002-0181-3412Skin cancerWhole slide imagesMultiple instance learningClass-incremental learningKnowledge distillationTEORÍA DE LA SEÑAL Y COMUNICACIONES[EN] Artificial intelligence (AI) agents encounter the problem of catastrophic forgetting when they are trained in sequentially with new data batches. This issue poses a barrier to the implementation of AI -based models in tasks that involve ongoing evolution, such as cancer prediction. Moreover, whole slide images (WSI) play a crucial role in cancer management, and their automated analysis has become increasingly popular in assisting pathologists during the diagnosis process. Incremental learning (IL) techniques aim to develop algorithms capable of retaining previously acquired information while also acquiring new insights to predict future data. Deep IL techniques need to address the challenges posed by the gigapixel scale of WSIs, which often necessitates the use of multiple instance learning (MIL) frameworks. In this paper, we introduce an IL algorithm tailored for analyzing WSIs within a MIL paradigm. The proposed M ultiple I nstance C lass- I ncremental L earning ( MICIL ) algorithm combines MIL with class-IL for the first time, allowing for the incremental prediction of multiple skin cancer subtypes from WSIs within a class-IL scenario. Our framework incorporates knowledge distillation and data rehearsal, along with a novel embedding-level distillation, aiming to preserve the latent space at the aggregated WSI level. Results demonstrate the algorithm's effectiveness in addressing the challenge of balancing IL-specific metrics, such as intransigence and forgetting, and solving the plasticity-stability dilemma.This work has received funding from Horizon 2020, the European Union's Framework Programme for Research and Innovation, under grant agreement No. 860627 (CLARIFY), the Spanish Ministry of Economy and Competitiveness through projects PID2019-105142RB-C21(AI4SKIN) and PID2022-140189OB-C21 (ASSIST), and GVA through the project INNEST/2021/321 (SAMUEL). The work of Rocio del Amorand Pablo Meseguer has been supported by the Spanish Ministry of Universities under an FPU Grant (FPU20/05263) and valgrAI - Valencian Graduate School and Research Network of Artificial Intelligence, respectively.ElsevierEscuela Técnica Superior de Ingeniería de TelecomunicaciónDepartamento de ComunicacionesInstituto Universitario de Investigación en Tecnología Centrada en el Ser HumanoEscuela Técnica Superior de Ingeniería IndustrialAgencia Estatal de Investigación• Agència Valenciana de la InnovacióEuropean CommissionUniversitat Politècnica de ValènciaMINISTERIO DE UNIVERSIDADES E INVESTIGACIONGeneralitat ValencianaRepositorio Institucional de la Universitat Politècnica de València Riunet20242024-06-01journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://riunet.upv.es/handle/10251/214706reponame:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valénciainstname:Universitat Politècnica de València (UPV)InglésengAgencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020 PID2019-105142RB-C21 CARACTERIZACION DE NEOPLASIAS DE CELULAS FUSIFORMES EN IMAGENES HISTOLOGICASAgencia Estatal de Investigación http://dx.doi.org/10.13039/501100011033 Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 PID2022-140189OB-C21 RECUPERACION DE IMAGENES BASADA EN EL CONTENIDO PARA EL DIAGNOSTICO DE TUMORES CUTANEOS PRIMARIOS Y SECUNDARIOSEuropean Commission https://doi.org/10.13039/501100000780 H2020 860627 CLoud ARtificial Intelligence For pathologYAgència Valenciana de la Innovació https://doi.org/10.13039/501100016028 INNEST%2F2021%2F321 SISTEMA DE VISIÓN ARTIFICIAL AUMENTADA PARA LA CARACTERIZACIÓN MOLECULAR Y MORFOLÓGICA DEL CÁNCER DE PIELMinisterio de Ciencia, Innovación y Universidades https://doi.org/10.13039/100014440 FPU20%2F05263 Sistema automático de clasificación de neoplasias cutáneas de células fusiformes basado en inteligencia artificialGeneralitat Valenciana https://doi.org/10.13039/501100003359 VALGRAI%2F22%2F4 Estrategias de aprendizaje profundo para el análisis de imágenes histopatológicasopen 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:10251/2147062026-06-13T07:49:27Z
score 15,811543

Similares en LA Referencia