Improved image registration by sparse patch-based deformation estimation

Minjeong Kim, Guorong Wu, Qian Wang, Seong Whan Lee, Dinggang Shen

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Despite intensive efforts for decades, deformable image registration is still a challenging problem due to the potential large anatomical differences across individual images, which limits the registration performance. Fortunately, this issue could be alleviated if a good initial deformation can be provided for the two images under registration, which are often termed as the moving subject and the fixed template, respectively. In this work, we present a novel patch-based initial deformation prediction framework for improving the performance of existing registration algorithms. Our main idea is to estimate the initial deformation between subject and template in a patch-wise fashion by using the sparse representation technique. We argue that two image patches should follow the same deformation toward the template image if their patch-wise appearance patterns are similar. To this end, our framework consists of two stages, i.e., the training stage and the application stage. In the training stage, we register all training images to the pre-selected template, such that the deformation of each training image with respect to the template is known. In the application stage, we apply the following four steps to efficiently calculate the initial deformation field for the new test subject: (1) We pick a small number of key points in the distinctive regions of the test subject; (2) for each key point, we extract a local patch and form a coupled appearance-deformation dictionary from training images where each dictionary atom consists of the image intensity patch as well as their respective local deformations; (3) a small set of training image patches in the coupled dictionary are selected to represent the image patch of each subject key point by sparse representation. Then, we can predict the initial deformation for each subject key point by propagating the pre-estimated deformations on the selected training patches with the same sparse representation coefficients; and (4) we employ thin-plate splines (TPS) to interpolate a dense initial deformation field by considering all key points as the control points. Thus, the conventional image registration problem becomes much easier in the sense that we only need to compute the remaining small deformation for completing the registration of the subject to the template. Experimental results on both simulated and real data show that the registration performance can be significantly improved after integrating our patch-based deformation prediction framework into the existing registration algorithms.

Original languageEnglish
Pages (from-to)257-268
Number of pages12
JournalNeuroImage
Volume105
DOIs
Publication statusPublished - 2015 Jan 5

Fingerprint

Individuality

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Improved image registration by sparse patch-based deformation estimation. / Kim, Minjeong; Wu, Guorong; Wang, Qian; Lee, Seong Whan; Shen, Dinggang.

In: NeuroImage, Vol. 105, 05.01.2015, p. 257-268.

Research output: Contribution to journalArticle

Kim, Minjeong ; Wu, Guorong ; Wang, Qian ; Lee, Seong Whan ; Shen, Dinggang. / Improved image registration by sparse patch-based deformation estimation. In: NeuroImage. 2015 ; Vol. 105. pp. 257-268.
@article{abb058ae8d00414db617ebfcb07e1a70,
title = "Improved image registration by sparse patch-based deformation estimation",
abstract = "Despite intensive efforts for decades, deformable image registration is still a challenging problem due to the potential large anatomical differences across individual images, which limits the registration performance. Fortunately, this issue could be alleviated if a good initial deformation can be provided for the two images under registration, which are often termed as the moving subject and the fixed template, respectively. In this work, we present a novel patch-based initial deformation prediction framework for improving the performance of existing registration algorithms. Our main idea is to estimate the initial deformation between subject and template in a patch-wise fashion by using the sparse representation technique. We argue that two image patches should follow the same deformation toward the template image if their patch-wise appearance patterns are similar. To this end, our framework consists of two stages, i.e., the training stage and the application stage. In the training stage, we register all training images to the pre-selected template, such that the deformation of each training image with respect to the template is known. In the application stage, we apply the following four steps to efficiently calculate the initial deformation field for the new test subject: (1) We pick a small number of key points in the distinctive regions of the test subject; (2) for each key point, we extract a local patch and form a coupled appearance-deformation dictionary from training images where each dictionary atom consists of the image intensity patch as well as their respective local deformations; (3) a small set of training image patches in the coupled dictionary are selected to represent the image patch of each subject key point by sparse representation. Then, we can predict the initial deformation for each subject key point by propagating the pre-estimated deformations on the selected training patches with the same sparse representation coefficients; and (4) we employ thin-plate splines (TPS) to interpolate a dense initial deformation field by considering all key points as the control points. Thus, the conventional image registration problem becomes much easier in the sense that we only need to compute the remaining small deformation for completing the registration of the subject to the template. Experimental results on both simulated and real data show that the registration performance can be significantly improved after integrating our patch-based deformation prediction framework into the existing registration algorithms.",
keywords = "Deformable image registration, Initial deformation prediction, Sparse representation",
author = "Minjeong Kim and Guorong Wu and Qian Wang and Lee, {Seong Whan} and Dinggang Shen",
year = "2015",
month = "1",
day = "5",
doi = "10.1016/j.neuroimage.2014.10.019",
language = "English",
volume = "105",
pages = "257--268",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Improved image registration by sparse patch-based deformation estimation

AU - Kim, Minjeong

AU - Wu, Guorong

AU - Wang, Qian

AU - Lee, Seong Whan

AU - Shen, Dinggang

PY - 2015/1/5

Y1 - 2015/1/5

N2 - Despite intensive efforts for decades, deformable image registration is still a challenging problem due to the potential large anatomical differences across individual images, which limits the registration performance. Fortunately, this issue could be alleviated if a good initial deformation can be provided for the two images under registration, which are often termed as the moving subject and the fixed template, respectively. In this work, we present a novel patch-based initial deformation prediction framework for improving the performance of existing registration algorithms. Our main idea is to estimate the initial deformation between subject and template in a patch-wise fashion by using the sparse representation technique. We argue that two image patches should follow the same deformation toward the template image if their patch-wise appearance patterns are similar. To this end, our framework consists of two stages, i.e., the training stage and the application stage. In the training stage, we register all training images to the pre-selected template, such that the deformation of each training image with respect to the template is known. In the application stage, we apply the following four steps to efficiently calculate the initial deformation field for the new test subject: (1) We pick a small number of key points in the distinctive regions of the test subject; (2) for each key point, we extract a local patch and form a coupled appearance-deformation dictionary from training images where each dictionary atom consists of the image intensity patch as well as their respective local deformations; (3) a small set of training image patches in the coupled dictionary are selected to represent the image patch of each subject key point by sparse representation. Then, we can predict the initial deformation for each subject key point by propagating the pre-estimated deformations on the selected training patches with the same sparse representation coefficients; and (4) we employ thin-plate splines (TPS) to interpolate a dense initial deformation field by considering all key points as the control points. Thus, the conventional image registration problem becomes much easier in the sense that we only need to compute the remaining small deformation for completing the registration of the subject to the template. Experimental results on both simulated and real data show that the registration performance can be significantly improved after integrating our patch-based deformation prediction framework into the existing registration algorithms.

AB - Despite intensive efforts for decades, deformable image registration is still a challenging problem due to the potential large anatomical differences across individual images, which limits the registration performance. Fortunately, this issue could be alleviated if a good initial deformation can be provided for the two images under registration, which are often termed as the moving subject and the fixed template, respectively. In this work, we present a novel patch-based initial deformation prediction framework for improving the performance of existing registration algorithms. Our main idea is to estimate the initial deformation between subject and template in a patch-wise fashion by using the sparse representation technique. We argue that two image patches should follow the same deformation toward the template image if their patch-wise appearance patterns are similar. To this end, our framework consists of two stages, i.e., the training stage and the application stage. In the training stage, we register all training images to the pre-selected template, such that the deformation of each training image with respect to the template is known. In the application stage, we apply the following four steps to efficiently calculate the initial deformation field for the new test subject: (1) We pick a small number of key points in the distinctive regions of the test subject; (2) for each key point, we extract a local patch and form a coupled appearance-deformation dictionary from training images where each dictionary atom consists of the image intensity patch as well as their respective local deformations; (3) a small set of training image patches in the coupled dictionary are selected to represent the image patch of each subject key point by sparse representation. Then, we can predict the initial deformation for each subject key point by propagating the pre-estimated deformations on the selected training patches with the same sparse representation coefficients; and (4) we employ thin-plate splines (TPS) to interpolate a dense initial deformation field by considering all key points as the control points. Thus, the conventional image registration problem becomes much easier in the sense that we only need to compute the remaining small deformation for completing the registration of the subject to the template. Experimental results on both simulated and real data show that the registration performance can be significantly improved after integrating our patch-based deformation prediction framework into the existing registration algorithms.

KW - Deformable image registration

KW - Initial deformation prediction

KW - Sparse representation

UR - http://www.scopus.com/inward/record.url?scp=84911369895&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84911369895&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2014.10.019

DO - 10.1016/j.neuroimage.2014.10.019

M3 - Article

C2 - 25451481

AN - SCOPUS:84911369895

VL - 105

SP - 257

EP - 268

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -