Weighted Shape-based Averaging with Neighborhood Prior Model for Multiple Atlas Fusion-based Medical Image Segmentation

Détails

Ressource 1Télécharger: BIB_524777162C71.P001.pdf (550.51 [Ko])
Etat: Serval
Version: Final published version
ID Serval
serval:BIB_524777162C71
Type
Article: article d'un périodique ou d'un magazine.
Collection
Publications
Titre
Weighted Shape-based Averaging with Neighborhood Prior Model for Multiple Atlas Fusion-based Medical Image Segmentation
Périodique
IEEE Signal Processing Letters
Auteur(s)
Gorthi S., Bach Cuadra M., Tercier P., Allal A.S., Thiran J.-Ph.
ISSN
1070-9908
Statut éditorial
Publié
Date de publication
11/2013
Peer-reviewed
Oui
Volume
20
Numéro
11
Pages
1036-1039
Langue
anglais
Résumé
In medical imaging, merging automated segmentations obtained from multiple atlases has become a standard practice for improving the accuracy. In this letter, we propose two new fusion methods: "Global Weighted Shape-Based Averaging" (GWSBA) and "Local Weighted Shape-Based Averaging" (LWSBA). These methods extend the well known Shape-Based Averaging (SBA) by additionally incorporating the similarity information between the reference (i.e., atlas) images and the target image to be segmented. We also propose a new spatially-varying similarity-weighted neighborhood prior model, and an edge-preserving smoothness term that can be used with many of the existing fusion methods. We first present our new Markov Random Field (MRF) based fusion framework that models the above mentioned information. The proposed methods are evaluated in the context of segmentation of lymph nodes in the head and neck 3D CT images, and they resulted in more accurate segmentations compared to the existing SBA.
Mots-clé
Biomedical imaging, Image segmentation, Signal processing, Atlas-based segmentation, MRF, SBA, Label fusion
Web of science
Création de la notice
29/08/2013 15:13
Dernière modification de la notice
03/03/2018 17:13
Données d'usage