COUPÉ, Pierrick; MANSENCAL, Boris; CLÉMENT, Michaël; GIRAUD, Rémi; DENIS DE SENNEVILLE, Baudouin; TA, Vinh-Thong; LEPETIT, Vincent; MANJÓN, José

doi:10.1016/j.neuroimage.2020.117026

Métadonnées

Licence d’utilisation du document

COUPÉ, Pierrick
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Patch-based processing for medical and natural images [PICTURA]

MANSENCAL, Boris
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

CLÉMENT, Michaël
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Patch-based processing for medical and natural images [PICTURA]

Langue

Article de revue

Ce document a été publié dans

NeuroImage. 2020-10, vol. 219, p. 117026

Elsevier

Résumé en anglais

Whole brain segmentation of fine-grained structures using deep learning (DL) is a very challenging task since the number of anatomical labels is very high compared to the number of available training images. To address this problem, previous DL methods proposed to use a single convolution neural network (CNN) or few independent CNNs. In this paper, we present a novel ensemble method based on a large number of CNNs processing different overlapping brain areas. Inspired by parliamentary decision-making systems, we propose a framework called AssemblyNet, made of two "assemblies" of U-Nets. Such a parliamentary system is capable of dealing with complex decisions, unseen problem and reaching a relevant consensus. AssemblyNet introduces sharing of knowledge among neighboring U-Nets, an "amendment" procedure made by the second assembly at higher-resolution to refine the decision taken by the first one, and a final decision obtained by majority voting. During our validation, AssemblyNet showed competitive performance compared to state-of-the-art methods such as U-Net, Joint label fusion and SLANT. Moreover, we investigated the scan-rescan consistency and the robustness to disease effects of our method. These experiences demonstrated the reliability of AssemblyNet. Finally, we showed the interest of using semi-supervised learning to improve the performance of our method.< Réduire

URI

https://oskar-bordeaux.fr/handle/20.500.12278/191883

DOI

http://dx.doi.org/10.1016/j.neuroimage.2020.117026

Project ANR

Apprentissage profond pour la volumétrie cérébrale : vers le BigData en neuroscience - ANR-18-CE45-0013
Translational Research and Advanced Imaging Laboratory - ANR-10-LABX-0057

Origine

Importé de hal

Unités de recherche

Institut de Mathématiques de Bordeaux (IMB) - UMR 5251