Statistically-constrained high-dimensional warping using wavelet-based priors

Zhong Xue, Dinggang Shen, Christos Davatzikos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, a Statistical Model of Deformation (SMD) that captures the statistical prior distribution of high-dimensional deformations more accurately and effectively than conventional PCA-based statistical shape models is used to regularize deformable registration. SMD utilizes a wavelet-based representation of statistical variation of a deformation field and its Jacobian, and it is able to capture both global and fine shape detail without overconstraining the deformation process. This approach is shown to produce more accurate and robust registration results in MR brain images, relative to the registration methods that use Laplacian-based smoothness constraints of deformation fields. In experiments, we evaluate the SMD-constrained registration by comparing the performance of registration with and without SMD in a specific deformable registration framework. The proposed method can potentially incorporate various registration algorithms to improve their robustness and stability using statistically-based regularization.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2006
DOIs
StatePublished - 2006
Externally publishedYes
Event2006 Conference on Computer Vision and Pattern Recognition Workshops - New York, NY, United States
Duration: Jun 17 2006Jun 22 2006

Other

Other2006 Conference on Computer Vision and Pattern Recognition Workshops
CountryUnited States
CityNew York, NY
Period6/17/066/22/06

Fingerprint

Brain
Statistical Models
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Xue, Z., Shen, D., & Davatzikos, C. (2006). Statistically-constrained high-dimensional warping using wavelet-based priors. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Vol. 2006). [1640512] https://doi.org/10.1109/CVPRW.2006.1

Statistically-constrained high-dimensional warping using wavelet-based priors. / Xue, Zhong; Shen, Dinggang; Davatzikos, Christos.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 2006 2006. 1640512.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xue, Z, Shen, D & Davatzikos, C 2006, Statistically-constrained high-dimensional warping using wavelet-based priors. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. vol. 2006, 1640512, 2006 Conference on Computer Vision and Pattern Recognition Workshops, New York, NY, United States, 6/17/06. https://doi.org/10.1109/CVPRW.2006.1
Xue Z, Shen D, Davatzikos C. Statistically-constrained high-dimensional warping using wavelet-based priors. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 2006. 2006. 1640512 https://doi.org/10.1109/CVPRW.2006.1
Xue, Zhong ; Shen, Dinggang ; Davatzikos, Christos. / Statistically-constrained high-dimensional warping using wavelet-based priors. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 2006 2006.
@inproceedings{7be264d4f7de4d7397f557e5b2b67982,
title = "Statistically-constrained high-dimensional warping using wavelet-based priors",
abstract = "In this paper, a Statistical Model of Deformation (SMD) that captures the statistical prior distribution of high-dimensional deformations more accurately and effectively than conventional PCA-based statistical shape models is used to regularize deformable registration. SMD utilizes a wavelet-based representation of statistical variation of a deformation field and its Jacobian, and it is able to capture both global and fine shape detail without overconstraining the deformation process. This approach is shown to produce more accurate and robust registration results in MR brain images, relative to the registration methods that use Laplacian-based smoothness constraints of deformation fields. In experiments, we evaluate the SMD-constrained registration by comparing the performance of registration with and without SMD in a specific deformable registration framework. The proposed method can potentially incorporate various registration algorithms to improve their robustness and stability using statistically-based regularization.",
author = "Zhong Xue and Dinggang Shen and Christos Davatzikos",
year = "2006",
doi = "10.1109/CVPRW.2006.1",
language = "English (US)",
isbn = "0769526462",
volume = "2006",
booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

}

TY - GEN

T1 - Statistically-constrained high-dimensional warping using wavelet-based priors

AU - Xue, Zhong

AU - Shen, Dinggang

AU - Davatzikos, Christos

PY - 2006

Y1 - 2006

N2 - In this paper, a Statistical Model of Deformation (SMD) that captures the statistical prior distribution of high-dimensional deformations more accurately and effectively than conventional PCA-based statistical shape models is used to regularize deformable registration. SMD utilizes a wavelet-based representation of statistical variation of a deformation field and its Jacobian, and it is able to capture both global and fine shape detail without overconstraining the deformation process. This approach is shown to produce more accurate and robust registration results in MR brain images, relative to the registration methods that use Laplacian-based smoothness constraints of deformation fields. In experiments, we evaluate the SMD-constrained registration by comparing the performance of registration with and without SMD in a specific deformable registration framework. The proposed method can potentially incorporate various registration algorithms to improve their robustness and stability using statistically-based regularization.

AB - In this paper, a Statistical Model of Deformation (SMD) that captures the statistical prior distribution of high-dimensional deformations more accurately and effectively than conventional PCA-based statistical shape models is used to regularize deformable registration. SMD utilizes a wavelet-based representation of statistical variation of a deformation field and its Jacobian, and it is able to capture both global and fine shape detail without overconstraining the deformation process. This approach is shown to produce more accurate and robust registration results in MR brain images, relative to the registration methods that use Laplacian-based smoothness constraints of deformation fields. In experiments, we evaluate the SMD-constrained registration by comparing the performance of registration with and without SMD in a specific deformable registration framework. The proposed method can potentially incorporate various registration algorithms to improve their robustness and stability using statistically-based regularization.

UR - http://www.scopus.com/inward/record.url?scp=33845548961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845548961&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2006.1

DO - 10.1109/CVPRW.2006.1

M3 - Conference contribution

SN - 0769526462

SN - 9780769526461

VL - 2006

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

ER -