Laplacian Forests: Semantic Image Segmentation by Guided Bagging

Abstract : This paper presents a new, efficient and accurate technique for the semantic segmentation of medical images. The paper builds upon the successful random decision forests model and improves on it by modifying the way in which randomness is injected into the tree training process. The contribution of this paper is two-fold. First, we replace the conventional bagging procedure (the uniform sampling of training images) with a guided bagging approach, which exploits the inherent structure and organization of the training image set. This allows the creation of decision trees that are specialized to a specific sub-type of images in the training set. Second, the segmentation of a previously unseen image happens via selection and application of only the trees that are relevant to the given test image. Tree selection is done automatically, via the learned image embedding, with more precisely a Laplacian eigenmap. We, therefore, call the proposed approach Laplacian Forests. We validate Laplacian Forests on a dataset of 256, manually segmented 3D CT scans of patients showing high variability in scanning protocols, resolution, body shape and anomalies. Compared with conventional decision forests, Laplacian Forests yield both higher training efficiency, due to the local analysis of the training image space, as well as higher segmentation accuracy, due to the specialization of the forest to image sub-types.
Type de document :
Communication dans un congrès
Polina Golland; Nobuhiko Hata; Christian Barillot; Joachim Hornegger; Robert Howe. MICCAI 2014 - 17th International Conference Medical Image Computing and Computer-Assisted Intervention, Sep 2014, Boston, United States. Springer, 8674, 2014, LNCS - Lecture Notes in Computer Science. 〈10.1007/978-3-319-10470-6_62〉
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01009672
Contributeur : Herve Lombaert <>
Soumis le : mardi 4 novembre 2014 - 10:48:26
Dernière modification le : jeudi 11 janvier 2018 - 16:40:59
Document(s) archivé(s) le : jeudi 5 février 2015 - 10:25:59

Fichier

MICCAI-LaplacianForest.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Herve Lombaert, Darko Zikic, Antonio Criminisi, Nicholas Ayache. Laplacian Forests: Semantic Image Segmentation by Guided Bagging. Polina Golland; Nobuhiko Hata; Christian Barillot; Joachim Hornegger; Robert Howe. MICCAI 2014 - 17th International Conference Medical Image Computing and Computer-Assisted Intervention, Sep 2014, Boston, United States. Springer, 8674, 2014, LNCS - Lecture Notes in Computer Science. 〈10.1007/978-3-319-10470-6_62〉. 〈hal-01009672〉

Partager

Métriques

Consultations de la notice

281

Téléchargements de fichiers

265