Spectral clustering to model deformations for fast multimodal prostate registration - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

Spectral clustering to model deformations for fast multimodal prostate registration

Résumé

This paper proposes a method to learn deformation parameters off-line for fast multimodal registration of ultrasound and magnetic resonance prostate images during ultrasound guided needle biopsy. The method is based on a learning phase where deformation models are built from the deformation parameters of a splinebased non-linear diffeomorphism between training ultrasound and magnetic resonance prostate images using spectral clustering. Deformation models comprising of the eigen-modes of each cluster in a Gaussian space are applied on a test magnetic resonance image to register with the test ultrasound prostate image. The deformation model with the least registration error is finally chosen as the optimal model for deformable registration. The rationale behind modeling deformations is to achieve fast multimodal registration of prostate images while maintaining registration accuracies which is otherwise computationally expensive. The method is validated for 25 patients each with a pair of corresponding magnetic resonance and ultrasound images in a leave-one-out validation framework. The average registration accuracies i.e. Dice similarity coefficient of 0.927 ± 0.025, 95% Hausdorff distance of 5.14 ± 3.67 mm and target registration error of 2.44±1.17 mm are obtained by our method with a speed-up in computation time by 98% when compared to Mitra et al. [7].
Fichier principal
Vignette du fichier
ICPR_Jhimli.pdf (148.32 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00710943 , version 1 (22-06-2012)

Identifiants

  • HAL Id : hal-00710943 , version 1

Citer

Jhimli Mitra, Zoltan Kato, Soumya Ghose, Désiré Sidibé, Robert Marti, et al.. Spectral clustering to model deformations for fast multimodal prostate registration. IAPR International Conference on Pattern Recognition, Nov 2012, Tsukba, Japan. ⟨hal-00710943⟩
139 Consultations
117 Téléchargements

Partager

Gmail Facebook X LinkedIn More