Attribute vector guided groupwise registration

Qian Wang, Guorong Wu, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Groupwise registration has been recently introduced to simultaneously register a group of images by avoiding the selection of a particular template. To achieve this, several methods have been proposed to take advantage of information-theoretic entropy measures based on image intensity. However, simplistic utilization of voxelwise image intensity is not sufficient to establish reliable correspondences, since it lacks important contextual information. Therefore, we explore the notion of attribute vector as the voxel signature, instead of image intensity, to guide the correspondence detection in groupwise registration. In particular, for each voxel, the attribute vector is computed from its multi-scale neighborhoods, in order to capture the geometric information at different scales. The probability density function (PDF) of each element in the attribute vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure in that local pattern. Eventually, with the help of Jensen-Shannon (JS) divergence, a group of subjects can be aligned simultaneously by minimizing the sum of JS divergences across the image domain and all attributes. We have employed our groupwise registration algorithm on both real (NIREP NA0 data set) and simulated data (12 pairs of normal control and simulated atrophic data set). The experimental results demonstrate that our method yields better registration accuracy, compared with a popular groupwise registration method.

Original languageEnglish
Pages (from-to)1485-1496
Number of pages12
JournalNeuroImage
Volume50
Issue number4
DOIs
Publication statusPublished - 2010 May 1
Externally publishedYes

Fingerprint

Entropy
Datasets

Keywords

  • Attribute vector
  • Groupwise registration
  • Jensen-Shannon divergence

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Attribute vector guided groupwise registration. / Wang, Qian; Wu, Guorong; Yap, Pew Thian; Shen, Dinggang.

In: NeuroImage, Vol. 50, No. 4, 01.05.2010, p. 1485-1496.

Research output: Contribution to journalArticle

Wang, Qian ; Wu, Guorong ; Yap, Pew Thian ; Shen, Dinggang. / Attribute vector guided groupwise registration. In: NeuroImage. 2010 ; Vol. 50, No. 4. pp. 1485-1496.
@article{ae06573b08844033a4b1c44d6ef2d5b9,
title = "Attribute vector guided groupwise registration",
abstract = "Groupwise registration has been recently introduced to simultaneously register a group of images by avoiding the selection of a particular template. To achieve this, several methods have been proposed to take advantage of information-theoretic entropy measures based on image intensity. However, simplistic utilization of voxelwise image intensity is not sufficient to establish reliable correspondences, since it lacks important contextual information. Therefore, we explore the notion of attribute vector as the voxel signature, instead of image intensity, to guide the correspondence detection in groupwise registration. In particular, for each voxel, the attribute vector is computed from its multi-scale neighborhoods, in order to capture the geometric information at different scales. The probability density function (PDF) of each element in the attribute vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure in that local pattern. Eventually, with the help of Jensen-Shannon (JS) divergence, a group of subjects can be aligned simultaneously by minimizing the sum of JS divergences across the image domain and all attributes. We have employed our groupwise registration algorithm on both real (NIREP NA0 data set) and simulated data (12 pairs of normal control and simulated atrophic data set). The experimental results demonstrate that our method yields better registration accuracy, compared with a popular groupwise registration method.",
keywords = "Attribute vector, Groupwise registration, Jensen-Shannon divergence",
author = "Qian Wang and Guorong Wu and Yap, {Pew Thian} and Dinggang Shen",
year = "2010",
month = "5",
day = "1",
doi = "10.1016/j.neuroimage.2010.01.040",
language = "English",
volume = "50",
pages = "1485--1496",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "4",

}

TY - JOUR

T1 - Attribute vector guided groupwise registration

AU - Wang, Qian

AU - Wu, Guorong

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2010/5/1

Y1 - 2010/5/1

N2 - Groupwise registration has been recently introduced to simultaneously register a group of images by avoiding the selection of a particular template. To achieve this, several methods have been proposed to take advantage of information-theoretic entropy measures based on image intensity. However, simplistic utilization of voxelwise image intensity is not sufficient to establish reliable correspondences, since it lacks important contextual information. Therefore, we explore the notion of attribute vector as the voxel signature, instead of image intensity, to guide the correspondence detection in groupwise registration. In particular, for each voxel, the attribute vector is computed from its multi-scale neighborhoods, in order to capture the geometric information at different scales. The probability density function (PDF) of each element in the attribute vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure in that local pattern. Eventually, with the help of Jensen-Shannon (JS) divergence, a group of subjects can be aligned simultaneously by minimizing the sum of JS divergences across the image domain and all attributes. We have employed our groupwise registration algorithm on both real (NIREP NA0 data set) and simulated data (12 pairs of normal control and simulated atrophic data set). The experimental results demonstrate that our method yields better registration accuracy, compared with a popular groupwise registration method.

AB - Groupwise registration has been recently introduced to simultaneously register a group of images by avoiding the selection of a particular template. To achieve this, several methods have been proposed to take advantage of information-theoretic entropy measures based on image intensity. However, simplistic utilization of voxelwise image intensity is not sufficient to establish reliable correspondences, since it lacks important contextual information. Therefore, we explore the notion of attribute vector as the voxel signature, instead of image intensity, to guide the correspondence detection in groupwise registration. In particular, for each voxel, the attribute vector is computed from its multi-scale neighborhoods, in order to capture the geometric information at different scales. The probability density function (PDF) of each element in the attribute vector is then estimated from the local neighborhood, providing a statistical summary of the underlying anatomical structure in that local pattern. Eventually, with the help of Jensen-Shannon (JS) divergence, a group of subjects can be aligned simultaneously by minimizing the sum of JS divergences across the image domain and all attributes. We have employed our groupwise registration algorithm on both real (NIREP NA0 data set) and simulated data (12 pairs of normal control and simulated atrophic data set). The experimental results demonstrate that our method yields better registration accuracy, compared with a popular groupwise registration method.

KW - Attribute vector

KW - Groupwise registration

KW - Jensen-Shannon divergence

UR - http://www.scopus.com/inward/record.url?scp=77950594481&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950594481&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2010.01.040

DO - 10.1016/j.neuroimage.2010.01.040

M3 - Article

C2 - 20097291

AN - SCOPUS:77950594481

VL - 50

SP - 1485

EP - 1496

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 4

ER -