A spline-based non-linear diffeomorphism for multimodal prostate registration

Jhimli Mitra, Z. Kato, Robert Martí, Arnau Oliver, Xavier Lladó, Désiré Sidibé, Soumya Ghose, Joan C. Vilanova, Josep Comet, Fabrice Meriaudeau

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

This paper presents a novel method for non-rigid registration of transrectal ultrasound and magnetic resonance prostate images based on a non-linear regularized framework of point correspondences obtained from a statistical measure of shape-contexts. The segmented prostate shapes are represented by shape-contexts and the Bhattacharyya distance between the shape representations is used to find the point correspondences between the 2D fixed and moving images. The registration method involves parametric estimation of the non-linear diffeomorphism between the multimodal images and has its basis in solving a set of non-linear equations of thin-plate splines. The solution is obtained as the least-squares solution of an over-determined system of non-linear equations constructed by integrating a set of non-linear functions over the fixed and moving images. However, this may not result in clinically acceptable transformations of the anatomical targets. Therefore, the regularized bending energy of the thin-plate splines along with the localization error of established correspondences should be included in the system of equations. The registration accuracies of the proposed method are evaluated in 20 pairs of prostate mid-gland ultrasound and magnetic resonance images. The results obtained in terms of Dice similarity coefficient show an average of 0.980 ± 0.004, average 95% Hausdorff distance of 1.63 ± 0.48. mm and mean target registration and target localization errors of 1.60 ± 1.17. mm and 0.15 ± 0.12. mm respectively.

Original languageEnglish
Pages (from-to)1259-1279
Number of pages21
JournalMedical Image Analysis
Volume16
Issue number6
DOIs
Publication statusPublished - Aug 2012

Fingerprint

Magnetic resonance
Nonlinear equations
Splines
Prostate
Ultrasonics
Magnetic Resonance Spectroscopy
Least-Squares Analysis

Keywords

  • Multimodal images
  • Non-linear registration
  • Prostate biopsy
  • Regularization
  • Thin-plate splines

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Radiology Nuclear Medicine and imaging
  • Health Informatics
  • Radiological and Ultrasound Technology

Cite this

Mitra, J., Kato, Z., Martí, R., Oliver, A., Lladó, X., Sidibé, D., ... Meriaudeau, F. (2012). A spline-based non-linear diffeomorphism for multimodal prostate registration. Medical Image Analysis, 16(6), 1259-1279. https://doi.org/10.1016/j.media.2012.04.006

A spline-based non-linear diffeomorphism for multimodal prostate registration. / Mitra, Jhimli; Kato, Z.; Martí, Robert; Oliver, Arnau; Lladó, Xavier; Sidibé, Désiré; Ghose, Soumya; Vilanova, Joan C.; Comet, Josep; Meriaudeau, Fabrice.

In: Medical Image Analysis, Vol. 16, No. 6, 08.2012, p. 1259-1279.

Research output: Contribution to journalArticle

Mitra, J, Kato, Z, Martí, R, Oliver, A, Lladó, X, Sidibé, D, Ghose, S, Vilanova, JC, Comet, J & Meriaudeau, F 2012, 'A spline-based non-linear diffeomorphism for multimodal prostate registration', Medical Image Analysis, vol. 16, no. 6, pp. 1259-1279. https://doi.org/10.1016/j.media.2012.04.006
Mitra, Jhimli ; Kato, Z. ; Martí, Robert ; Oliver, Arnau ; Lladó, Xavier ; Sidibé, Désiré ; Ghose, Soumya ; Vilanova, Joan C. ; Comet, Josep ; Meriaudeau, Fabrice. / A spline-based non-linear diffeomorphism for multimodal prostate registration. In: Medical Image Analysis. 2012 ; Vol. 16, No. 6. pp. 1259-1279.
@article{9f29ef50829845d184c01d461818027c,
title = "A spline-based non-linear diffeomorphism for multimodal prostate registration",
abstract = "This paper presents a novel method for non-rigid registration of transrectal ultrasound and magnetic resonance prostate images based on a non-linear regularized framework of point correspondences obtained from a statistical measure of shape-contexts. The segmented prostate shapes are represented by shape-contexts and the Bhattacharyya distance between the shape representations is used to find the point correspondences between the 2D fixed and moving images. The registration method involves parametric estimation of the non-linear diffeomorphism between the multimodal images and has its basis in solving a set of non-linear equations of thin-plate splines. The solution is obtained as the least-squares solution of an over-determined system of non-linear equations constructed by integrating a set of non-linear functions over the fixed and moving images. However, this may not result in clinically acceptable transformations of the anatomical targets. Therefore, the regularized bending energy of the thin-plate splines along with the localization error of established correspondences should be included in the system of equations. The registration accuracies of the proposed method are evaluated in 20 pairs of prostate mid-gland ultrasound and magnetic resonance images. The results obtained in terms of Dice similarity coefficient show an average of 0.980 ± 0.004, average 95{\%} Hausdorff distance of 1.63 ± 0.48. mm and mean target registration and target localization errors of 1.60 ± 1.17. mm and 0.15 ± 0.12. mm respectively.",
keywords = "Multimodal images, Non-linear registration, Prostate biopsy, Regularization, Thin-plate splines",
author = "Jhimli Mitra and Z. Kato and Robert Mart{\'i} and Arnau Oliver and Xavier Llad{\'o} and D{\'e}sir{\'e} Sidib{\'e} and Soumya Ghose and Vilanova, {Joan C.} and Josep Comet and Fabrice Meriaudeau",
year = "2012",
month = "8",
doi = "10.1016/j.media.2012.04.006",
language = "English",
volume = "16",
pages = "1259--1279",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "6",

}

TY - JOUR

T1 - A spline-based non-linear diffeomorphism for multimodal prostate registration

AU - Mitra, Jhimli

AU - Kato, Z.

AU - Martí, Robert

AU - Oliver, Arnau

AU - Lladó, Xavier

AU - Sidibé, Désiré

AU - Ghose, Soumya

AU - Vilanova, Joan C.

AU - Comet, Josep

AU - Meriaudeau, Fabrice

PY - 2012/8

Y1 - 2012/8

N2 - This paper presents a novel method for non-rigid registration of transrectal ultrasound and magnetic resonance prostate images based on a non-linear regularized framework of point correspondences obtained from a statistical measure of shape-contexts. The segmented prostate shapes are represented by shape-contexts and the Bhattacharyya distance between the shape representations is used to find the point correspondences between the 2D fixed and moving images. The registration method involves parametric estimation of the non-linear diffeomorphism between the multimodal images and has its basis in solving a set of non-linear equations of thin-plate splines. The solution is obtained as the least-squares solution of an over-determined system of non-linear equations constructed by integrating a set of non-linear functions over the fixed and moving images. However, this may not result in clinically acceptable transformations of the anatomical targets. Therefore, the regularized bending energy of the thin-plate splines along with the localization error of established correspondences should be included in the system of equations. The registration accuracies of the proposed method are evaluated in 20 pairs of prostate mid-gland ultrasound and magnetic resonance images. The results obtained in terms of Dice similarity coefficient show an average of 0.980 ± 0.004, average 95% Hausdorff distance of 1.63 ± 0.48. mm and mean target registration and target localization errors of 1.60 ± 1.17. mm and 0.15 ± 0.12. mm respectively.

AB - This paper presents a novel method for non-rigid registration of transrectal ultrasound and magnetic resonance prostate images based on a non-linear regularized framework of point correspondences obtained from a statistical measure of shape-contexts. The segmented prostate shapes are represented by shape-contexts and the Bhattacharyya distance between the shape representations is used to find the point correspondences between the 2D fixed and moving images. The registration method involves parametric estimation of the non-linear diffeomorphism between the multimodal images and has its basis in solving a set of non-linear equations of thin-plate splines. The solution is obtained as the least-squares solution of an over-determined system of non-linear equations constructed by integrating a set of non-linear functions over the fixed and moving images. However, this may not result in clinically acceptable transformations of the anatomical targets. Therefore, the regularized bending energy of the thin-plate splines along with the localization error of established correspondences should be included in the system of equations. The registration accuracies of the proposed method are evaluated in 20 pairs of prostate mid-gland ultrasound and magnetic resonance images. The results obtained in terms of Dice similarity coefficient show an average of 0.980 ± 0.004, average 95% Hausdorff distance of 1.63 ± 0.48. mm and mean target registration and target localization errors of 1.60 ± 1.17. mm and 0.15 ± 0.12. mm respectively.

KW - Multimodal images

KW - Non-linear registration

KW - Prostate biopsy

KW - Regularization

KW - Thin-plate splines

UR - http://www.scopus.com/inward/record.url?scp=84866118888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866118888&partnerID=8YFLogxK

U2 - 10.1016/j.media.2012.04.006

DO - 10.1016/j.media.2012.04.006

M3 - Article

C2 - 22705289

AN - SCOPUS:84866118888

VL - 16

SP - 1259

EP - 1279

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 6

ER -