A Markov random field approach for topology-preserving registration: application to object-based tomographic image interpolation

Lucilio Cordero-Grande, Gonzalo Vegas-Sanchez-Ferrero, Juan Pablo Casaseca, Carlos Alberola-Lopez

Research output: Contribution to journalArticle

Abstract

This paper proposes a topology-preserving multiresolution elastic registration method based on a discrete Markov random field of deformations and a block-matching procedure. The method is applied to the object-based interpolation of tomographic slices. For that purpose, the fidelity of a given deformation to the data is established by a block-matching strategy based on intensity-and gradient-related features, the smoothness of the transformation is favored by an appropriate prior on the field, and the deformation is guaranteed to maintain the topology by imposing some hard constraints on the local configurations of the field. The resulting deformation is defined as the maximum a posteriori configuration. Additionally, the relative influence of the fidelity and smoothness terms is weighted by the unsupervised estimation of the field parameters. In order to obtain an unbiased interpolation result, the registration is performed both in the forward and backward directions, and the resulting transformations are combined by using the local information content of the deformation. The method is applied to magnetic resonance and computed tomography acquisitions of the brain and the torso. Quantitative comparisons offer an overall improvement in performance with respect to related works in the literature. Additionally, the application of the interpolation method to cardiac magnetic resonance images has shown that the removal of any of the main components of the algorithm results in a decrease in performance which has proven to be statistically significant.
Original languageEnglish
Pages (from-to)2047-2061
Number of pages15
JournalIEEE Transactions on Image Processing
Volume21
Issue number4
DOIs
Publication statusPublished - Apr 2012
Externally publishedYes

Keywords

  • Markov random field (MRF)
  • parameter estimation
  • tomography interpolation
  • topology-preserving registration

Cite this

Cordero-Grande, Lucilio ; Vegas-Sanchez-Ferrero, Gonzalo ; Casaseca, Juan Pablo ; Alberola-Lopez, Carlos. / A Markov random field approach for topology-preserving registration : application to object-based tomographic image interpolation. In: IEEE Transactions on Image Processing. 2012 ; Vol. 21, No. 4. pp. 2047-2061.
@article{5b3582f316114db7997dc93f5bd4fc28,
title = "A Markov random field approach for topology-preserving registration: application to object-based tomographic image interpolation",
abstract = "This paper proposes a topology-preserving multiresolution elastic registration method based on a discrete Markov random field of deformations and a block-matching procedure. The method is applied to the object-based interpolation of tomographic slices. For that purpose, the fidelity of a given deformation to the data is established by a block-matching strategy based on intensity-and gradient-related features, the smoothness of the transformation is favored by an appropriate prior on the field, and the deformation is guaranteed to maintain the topology by imposing some hard constraints on the local configurations of the field. The resulting deformation is defined as the maximum a posteriori configuration. Additionally, the relative influence of the fidelity and smoothness terms is weighted by the unsupervised estimation of the field parameters. In order to obtain an unbiased interpolation result, the registration is performed both in the forward and backward directions, and the resulting transformations are combined by using the local information content of the deformation. The method is applied to magnetic resonance and computed tomography acquisitions of the brain and the torso. Quantitative comparisons offer an overall improvement in performance with respect to related works in the literature. Additionally, the application of the interpolation method to cardiac magnetic resonance images has shown that the removal of any of the main components of the algorithm results in a decrease in performance which has proven to be statistically significant.",
keywords = "Markov random field (MRF), parameter estimation, tomography interpolation, topology-preserving registration",
author = "Lucilio Cordero-Grande and Gonzalo Vegas-Sanchez-Ferrero and Casaseca, {Juan Pablo} and Carlos Alberola-Lopez",
year = "2012",
month = "4",
doi = "10.1109/TIP.2011.2171354",
language = "English",
volume = "21",
pages = "2047--2061",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "IEEE",
number = "4",

}

A Markov random field approach for topology-preserving registration : application to object-based tomographic image interpolation. / Cordero-Grande, Lucilio; Vegas-Sanchez-Ferrero, Gonzalo; Casaseca, Juan Pablo; Alberola-Lopez, Carlos.

In: IEEE Transactions on Image Processing, Vol. 21, No. 4, 04.2012, p. 2047-2061.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Markov random field approach for topology-preserving registration

T2 - application to object-based tomographic image interpolation

AU - Cordero-Grande, Lucilio

AU - Vegas-Sanchez-Ferrero, Gonzalo

AU - Casaseca, Juan Pablo

AU - Alberola-Lopez, Carlos

PY - 2012/4

Y1 - 2012/4

N2 - This paper proposes a topology-preserving multiresolution elastic registration method based on a discrete Markov random field of deformations and a block-matching procedure. The method is applied to the object-based interpolation of tomographic slices. For that purpose, the fidelity of a given deformation to the data is established by a block-matching strategy based on intensity-and gradient-related features, the smoothness of the transformation is favored by an appropriate prior on the field, and the deformation is guaranteed to maintain the topology by imposing some hard constraints on the local configurations of the field. The resulting deformation is defined as the maximum a posteriori configuration. Additionally, the relative influence of the fidelity and smoothness terms is weighted by the unsupervised estimation of the field parameters. In order to obtain an unbiased interpolation result, the registration is performed both in the forward and backward directions, and the resulting transformations are combined by using the local information content of the deformation. The method is applied to magnetic resonance and computed tomography acquisitions of the brain and the torso. Quantitative comparisons offer an overall improvement in performance with respect to related works in the literature. Additionally, the application of the interpolation method to cardiac magnetic resonance images has shown that the removal of any of the main components of the algorithm results in a decrease in performance which has proven to be statistically significant.

AB - This paper proposes a topology-preserving multiresolution elastic registration method based on a discrete Markov random field of deformations and a block-matching procedure. The method is applied to the object-based interpolation of tomographic slices. For that purpose, the fidelity of a given deformation to the data is established by a block-matching strategy based on intensity-and gradient-related features, the smoothness of the transformation is favored by an appropriate prior on the field, and the deformation is guaranteed to maintain the topology by imposing some hard constraints on the local configurations of the field. The resulting deformation is defined as the maximum a posteriori configuration. Additionally, the relative influence of the fidelity and smoothness terms is weighted by the unsupervised estimation of the field parameters. In order to obtain an unbiased interpolation result, the registration is performed both in the forward and backward directions, and the resulting transformations are combined by using the local information content of the deformation. The method is applied to magnetic resonance and computed tomography acquisitions of the brain and the torso. Quantitative comparisons offer an overall improvement in performance with respect to related works in the literature. Additionally, the application of the interpolation method to cardiac magnetic resonance images has shown that the removal of any of the main components of the algorithm results in a decrease in performance which has proven to be statistically significant.

KW - Markov random field (MRF)

KW - parameter estimation

KW - tomography interpolation

KW - topology-preserving registration

U2 - 10.1109/TIP.2011.2171354

DO - 10.1109/TIP.2011.2171354

M3 - Article

VL - 21

SP - 2047

EP - 2061

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 4

ER -