Mostrar el registro sencillo del ítem
dc.contributor.author | Kamraoui, Reda Abdellah | es_ES |
dc.contributor.author | Ta, Vinh-Thong | es_ES |
dc.contributor.author | Tourdias, Thomas | es_ES |
dc.contributor.author | Mansencal, Boris | es_ES |
dc.contributor.author | Manjón Herrera, José Vicente | es_ES |
dc.contributor.author | Coupé, Pierrick | es_ES |
dc.date.accessioned | 2023-11-13T19:03:18Z | |
dc.date.available | 2023-11-13T19:03:18Z | |
dc.date.issued | 2022-02 | es_ES |
dc.identifier.issn | 1361-8415 | es_ES |
dc.identifier.uri | http://hdl.handle.net/10251/199575 | |
dc.description.abstract | [EN] Recently, segmentation methods based on Convolutional Neural Networks (CNNs) showed promising per-formance in automatic Multiple Sclerosis (MS) lesions segmentation. These techniques have even out-performed human experts in controlled evaluation conditions such as Longitudinal MS Lesion Segmenta-tion Challenge (ISBI Challenge). However, state-of-the-art approaches trained to perform well on highly-controlled datasets fail to generalize on clinical data from unseen datasets. Instead of proposing another improvement of the segmentation accuracy, we propose a novel method robust to domain shift and per-forming well on unseen datasets, called DeepLesionBrain (DLB). This generalization property results from three main contributions. First, DLB is based on a large group of compact 3D CNNs. This spatially dis-tributed strategy aims to produce a robust prediction despite the risk of generalization failure of some individual networks. Second, we propose a hierarchical specialization learning (HSL) by pre-training a generic network over the whole brain, before using its weights as initialization to locally specialized net-works. By this end, DLB learns both generic features extracted at global image level and specific features extracted at local image level. Finally, DLB includes a new image quality data augmentation to reduce dependency to training data specificity (e.g., acquisition protocol). DLB generalization was validated in cross-dataset experiments on MSSEG'16, ISBI challenge, and in-house datasets. During experiments, DLB showed higher segmentation accuracy, better segmentation consistency and greater generalization per-formance compared to state-of-the-art methods. Therefore, DLB offers a robust framework well-suited for clinical practice. (c) 2021 Published by Elsevier B.V. | es_ES |
dc.description.sponsorship | This work benefited from the support of the project Deep-volBrain of the French National Research Agency (ANR-18-CE45-0013) . This study was achieved within the context of the Labo-ratory of Excellence TRAIL ANR-10-LABX-57 for the BigDataBrain project. Moreover, we thank the Investments for the future Program IdEx Bordeaux (ANR-10-IDEX-03-02, HL-MRI Project) , Clus-ter of excellence CPU and the CNRS/INSERM for the DeepMulti-Brain project. This study has also been supported by the DPI2017-87743-R grant from the Spanish Ministerio de Economia, Industria Competitividad. The authors gratefully acknowledge the support of NVIDIA Corporation with their donation of the TITAN Xp GPU used in this research. | es_ES |
dc.language | Inglés | es_ES |
dc.publisher | Elsevier | es_ES |
dc.relation.ispartof | Medical Image Analysis | es_ES |
dc.rights | Reserva de todos los derechos | es_ES |
dc.subject | Multiple Sclerosis Segmentation | es_ES |
dc.subject | Deep Learning | es_ES |
dc.subject | Domain Generalization | es_ES |
dc.subject | MRI | es_ES |
dc.subject | Segmentation | es_ES |
dc.subject.classification | FISICA APLICADA | es_ES |
dc.title | DeepLesionBrain: Towards a broader deep-learning generalization for multiple sclerosis lesion segmentation | es_ES |
dc.type | Artículo | es_ES |
dc.identifier.doi | 10.1016/j.media.2021.102312 | es_ES |
dc.relation.projectID | info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016/DPI2017-87743-R/ES/DESARROLLO DE UNA PLATAFORMA ONLINE PARA EL ANALISIS ANATOMICO DEL CEREBRO TOLERANTE A LA PRESENCIA DE ALTERACIONES PATOLOGICAS/ | es_ES |
dc.relation.projectID | info:eu-repo/grantAgreement/ANR//ANR-10-LABX-57/ | es_ES |
dc.relation.projectID | info:eu-repo/grantAgreement/ANR//ANR-18-CE45-0013/ | es_ES |
dc.relation.projectID | info:eu-repo/grantAgreement/Université de Bordeaux//ANR-10-IDEX-03-02//Initiative of Excellence/ | es_ES |
dc.rights.accessRights | Cerrado | es_ES |
dc.contributor.affiliation | Universitat Politècnica de València. Escola Tècnica Superior d'Enginyeria Informàtica | es_ES |
dc.description.bibliographicCitation | Kamraoui, RA.; Ta, V.; Tourdias, T.; Mansencal, B.; Manjón Herrera, JV.; Coupé, P. (2022). DeepLesionBrain: Towards a broader deep-learning generalization for multiple sclerosis lesion segmentation. Medical Image Analysis. 76:1-13. https://doi.org/10.1016/j.media.2021.102312 | es_ES |
dc.description.accrualMethod | S | es_ES |
dc.relation.publisherversion | https://doi.org/10.1016/j.media.2021.102312 | es_ES |
dc.description.upvformatpinicio | 1 | es_ES |
dc.description.upvformatpfin | 13 | es_ES |
dc.type.version | info:eu-repo/semantics/publishedVersion | es_ES |
dc.description.volume | 76 | es_ES |
dc.identifier.pmid | 34894571 | es_ES |
dc.relation.pasarela | S\484083 | es_ES |
dc.contributor.funder | Nvidia | es_ES |
dc.contributor.funder | Université de Bordeaux | es_ES |
dc.contributor.funder | Agencia Estatal de Investigación | es_ES |
dc.contributor.funder | Agence Nationale de la Recherche, Francia | es_ES |