Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework

J. S. Kim, J. M. Lee, J. J. Kim, B. Y. Choe, Chil Hwan Oh, S. H. Nam, J. S. Kwon, Sun I. Kim

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The objective of this work was to provide a new, precise registration of the cortical mantle with a non-linear transformation. Image registration is broadly classified into two types, using intensity similarity and feature similarity. Whereas the former approach has merit in global brain matching, the latter provides a fast registration centred on a region of interest. The hybrid registration proposed in this paper was achieved using a Bayesian framework, which consisted of a likelihood model including intensity similarity and a prior model including feature information and a smoothing constraint. In this approach, each voxel was spatially transformed, so that the distance between corresponding features was shortened and also so that the intensity correlation was maximised. The result of the hybrid method clearly showed a good match of global brain (r=0.930) by including intensity similarity. Moreover, this method compensated for the approximated sulcus of the feature-based method with intensity information, so that the geometric shape and thickness of the sulcus at the feature-defined region was likely to be registered. The accuracy in the feature-defined area was improved by 33.4% and 7.5% compared with feature-based and intensity-based methods, respectively.

Original languageEnglish
Pages (from-to)473-480
Number of pages8
JournalMedical and Biological Engineering and Computing
Volume41
Issue number4
Publication statusPublished - 2003 Jul 1

Fingerprint

Brain
Image registration

Keywords

  • Bayesian framework
  • Feature information
  • Image registration
  • Intensity similarity

ASJC Scopus subject areas

  • Biomedical Engineering
  • Computer Science Applications

Cite this

Kim, J. S., Lee, J. M., Kim, J. J., Choe, B. Y., Oh, C. H., Nam, S. H., ... Kim, S. I. (2003). Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework. Medical and Biological Engineering and Computing, 41(4), 473-480.

Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework. / Kim, J. S.; Lee, J. M.; Kim, J. J.; Choe, B. Y.; Oh, Chil Hwan; Nam, S. H.; Kwon, J. S.; Kim, Sun I.

In: Medical and Biological Engineering and Computing, Vol. 41, No. 4, 01.07.2003, p. 473-480.

Research output: Contribution to journalArticle

Kim, JS, Lee, JM, Kim, JJ, Choe, BY, Oh, CH, Nam, SH, Kwon, JS & Kim, SI 2003, 'Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework', Medical and Biological Engineering and Computing, vol. 41, no. 4, pp. 473-480.
Kim, J. S. ; Lee, J. M. ; Kim, J. J. ; Choe, B. Y. ; Oh, Chil Hwan ; Nam, S. H. ; Kwon, J. S. ; Kim, Sun I. / Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework. In: Medical and Biological Engineering and Computing. 2003 ; Vol. 41, No. 4. pp. 473-480.
@article{a7de75695e944604901218080f5c7df4,
title = "Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework",
abstract = "The objective of this work was to provide a new, precise registration of the cortical mantle with a non-linear transformation. Image registration is broadly classified into two types, using intensity similarity and feature similarity. Whereas the former approach has merit in global brain matching, the latter provides a fast registration centred on a region of interest. The hybrid registration proposed in this paper was achieved using a Bayesian framework, which consisted of a likelihood model including intensity similarity and a prior model including feature information and a smoothing constraint. In this approach, each voxel was spatially transformed, so that the distance between corresponding features was shortened and also so that the intensity correlation was maximised. The result of the hybrid method clearly showed a good match of global brain (r=0.930) by including intensity similarity. Moreover, this method compensated for the approximated sulcus of the feature-based method with intensity information, so that the geometric shape and thickness of the sulcus at the feature-defined region was likely to be registered. The accuracy in the feature-defined area was improved by 33.4{\%} and 7.5{\%} compared with feature-based and intensity-based methods, respectively.",
keywords = "Bayesian framework, Feature information, Image registration, Intensity similarity",
author = "Kim, {J. S.} and Lee, {J. M.} and Kim, {J. J.} and Choe, {B. Y.} and Oh, {Chil Hwan} and Nam, {S. H.} and Kwon, {J. S.} and Kim, {Sun I.}",
year = "2003",
month = "7",
day = "1",
language = "English",
volume = "41",
pages = "473--480",
journal = "Medical and Biological Engineering and Computing",
issn = "0140-0118",
publisher = "Springer Verlag",
number = "4",

}

TY - JOUR

T1 - Non-linear registration for brain images by maximising feature and intensity similarities with a Bayesian framework

AU - Kim, J. S.

AU - Lee, J. M.

AU - Kim, J. J.

AU - Choe, B. Y.

AU - Oh, Chil Hwan

AU - Nam, S. H.

AU - Kwon, J. S.

AU - Kim, Sun I.

PY - 2003/7/1

Y1 - 2003/7/1

N2 - The objective of this work was to provide a new, precise registration of the cortical mantle with a non-linear transformation. Image registration is broadly classified into two types, using intensity similarity and feature similarity. Whereas the former approach has merit in global brain matching, the latter provides a fast registration centred on a region of interest. The hybrid registration proposed in this paper was achieved using a Bayesian framework, which consisted of a likelihood model including intensity similarity and a prior model including feature information and a smoothing constraint. In this approach, each voxel was spatially transformed, so that the distance between corresponding features was shortened and also so that the intensity correlation was maximised. The result of the hybrid method clearly showed a good match of global brain (r=0.930) by including intensity similarity. Moreover, this method compensated for the approximated sulcus of the feature-based method with intensity information, so that the geometric shape and thickness of the sulcus at the feature-defined region was likely to be registered. The accuracy in the feature-defined area was improved by 33.4% and 7.5% compared with feature-based and intensity-based methods, respectively.

AB - The objective of this work was to provide a new, precise registration of the cortical mantle with a non-linear transformation. Image registration is broadly classified into two types, using intensity similarity and feature similarity. Whereas the former approach has merit in global brain matching, the latter provides a fast registration centred on a region of interest. The hybrid registration proposed in this paper was achieved using a Bayesian framework, which consisted of a likelihood model including intensity similarity and a prior model including feature information and a smoothing constraint. In this approach, each voxel was spatially transformed, so that the distance between corresponding features was shortened and also so that the intensity correlation was maximised. The result of the hybrid method clearly showed a good match of global brain (r=0.930) by including intensity similarity. Moreover, this method compensated for the approximated sulcus of the feature-based method with intensity information, so that the geometric shape and thickness of the sulcus at the feature-defined region was likely to be registered. The accuracy in the feature-defined area was improved by 33.4% and 7.5% compared with feature-based and intensity-based methods, respectively.

KW - Bayesian framework

KW - Feature information

KW - Image registration

KW - Intensity similarity

UR - http://www.scopus.com/inward/record.url?scp=0041807935&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041807935&partnerID=8YFLogxK

M3 - Article

C2 - 12892372

AN - SCOPUS:0041807935

VL - 41

SP - 473

EP - 480

JO - Medical and Biological Engineering and Computing

JF - Medical and Biological Engineering and Computing

SN - 0140-0118

IS - 4

ER -