A novel longitudinal atlas construction framework by groupwise registration of subject image sequences

Shu Liao, Hongjun Jia, Guorong Wu, Dinggang Shen

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

Abstract

Longitudinal atlas construction is a challenging task in medical image analysis. Given a set of longitudinal images of different subjects, the task is how to construct the unbias longitudinal atlas sequence reflecting the anatomical changes over time. In this paper, a novel longitudinal atlas construction framework is proposed. The main contributions of the proposed method lie in the following aspects: (1) Subject-specific longitudinal information is captured by establishing a robust growth model for each subject. (2) The trajectory constraints are enforced for both subject image sequences and the atlas sequence, and only one transformation is needed for each subject to map its image sequence to the atlas sequence while preserving the temporal correspondence. (3) The longitudinal atlases are estimated by groupwise registration and kernel regression, thus no explicit template is used and the atlases are constructed without introducing bias due to the selection of the explicit template. (4) The proposed method is general, where the number of longitudinal images of each subject and the time points at which the images are taken can be different. The proposed method is evaluated on a longitudinal database and compared with a state-of-the-art longitudinal atlas construction method. Experimental results show that the proposed method achieves more consistent spatial-temporal correspondence as well as higher registration accuracy than the compared method.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages283-295
Number of pages13
Volume6801 LNCS
DOIs
Publication statusPublished - 2011 Jun 30
Externally publishedYes
Event22nd International Conference on Information Processing in Medical Imaging, IPMI 2011 - Kloster Irsee, Germany
Duration: 2011 Jul 32011 Jul 8

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6801 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other22nd International Conference on Information Processing in Medical Imaging, IPMI 2011
CountryGermany
CityKloster Irsee
Period11/7/311/7/8

Fingerprint

Atlas
Image Sequence
Registration
Image analysis
Trajectories
Template
Correspondence
Medical Image Analysis
Framework
Kernel Regression
Growth Model
Trajectory
Experimental Results

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Liao, S., Jia, H., Wu, G., & Shen, D. (2011). A novel longitudinal atlas construction framework by groupwise registration of subject image sequences. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6801 LNCS, pp. 283-295). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6801 LNCS). https://doi.org/10.1007/978-3-642-22092-0_24

A novel longitudinal atlas construction framework by groupwise registration of subject image sequences. / Liao, Shu; Jia, Hongjun; Wu, Guorong; Shen, Dinggang.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6801 LNCS 2011. p. 283-295 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6801 LNCS).

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

Liao, S, Jia, H, Wu, G & Shen, D 2011, A novel longitudinal atlas construction framework by groupwise registration of subject image sequences. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 6801 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6801 LNCS, pp. 283-295, 22nd International Conference on Information Processing in Medical Imaging, IPMI 2011, Kloster Irsee, Germany, 11/7/3. https://doi.org/10.1007/978-3-642-22092-0_24
Liao S, Jia H, Wu G, Shen D. A novel longitudinal atlas construction framework by groupwise registration of subject image sequences. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6801 LNCS. 2011. p. 283-295. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-22092-0_24
Liao, Shu ; Jia, Hongjun ; Wu, Guorong ; Shen, Dinggang. / A novel longitudinal atlas construction framework by groupwise registration of subject image sequences. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6801 LNCS 2011. pp. 283-295 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{47e3f67941254ecc8481220760c6e48e,
title = "A novel longitudinal atlas construction framework by groupwise registration of subject image sequences",
abstract = "Longitudinal atlas construction is a challenging task in medical image analysis. Given a set of longitudinal images of different subjects, the task is how to construct the unbias longitudinal atlas sequence reflecting the anatomical changes over time. In this paper, a novel longitudinal atlas construction framework is proposed. The main contributions of the proposed method lie in the following aspects: (1) Subject-specific longitudinal information is captured by establishing a robust growth model for each subject. (2) The trajectory constraints are enforced for both subject image sequences and the atlas sequence, and only one transformation is needed for each subject to map its image sequence to the atlas sequence while preserving the temporal correspondence. (3) The longitudinal atlases are estimated by groupwise registration and kernel regression, thus no explicit template is used and the atlases are constructed without introducing bias due to the selection of the explicit template. (4) The proposed method is general, where the number of longitudinal images of each subject and the time points at which the images are taken can be different. The proposed method is evaluated on a longitudinal database and compared with a state-of-the-art longitudinal atlas construction method. Experimental results show that the proposed method achieves more consistent spatial-temporal correspondence as well as higher registration accuracy than the compared method.",
author = "Shu Liao and Hongjun Jia and Guorong Wu and Dinggang Shen",
year = "2011",
month = "6",
day = "30",
doi = "10.1007/978-3-642-22092-0_24",
language = "English",
isbn = "9783642220913",
volume = "6801 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "283--295",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - A novel longitudinal atlas construction framework by groupwise registration of subject image sequences

AU - Liao, Shu

AU - Jia, Hongjun

AU - Wu, Guorong

AU - Shen, Dinggang

PY - 2011/6/30

Y1 - 2011/6/30

N2 - Longitudinal atlas construction is a challenging task in medical image analysis. Given a set of longitudinal images of different subjects, the task is how to construct the unbias longitudinal atlas sequence reflecting the anatomical changes over time. In this paper, a novel longitudinal atlas construction framework is proposed. The main contributions of the proposed method lie in the following aspects: (1) Subject-specific longitudinal information is captured by establishing a robust growth model for each subject. (2) The trajectory constraints are enforced for both subject image sequences and the atlas sequence, and only one transformation is needed for each subject to map its image sequence to the atlas sequence while preserving the temporal correspondence. (3) The longitudinal atlases are estimated by groupwise registration and kernel regression, thus no explicit template is used and the atlases are constructed without introducing bias due to the selection of the explicit template. (4) The proposed method is general, where the number of longitudinal images of each subject and the time points at which the images are taken can be different. The proposed method is evaluated on a longitudinal database and compared with a state-of-the-art longitudinal atlas construction method. Experimental results show that the proposed method achieves more consistent spatial-temporal correspondence as well as higher registration accuracy than the compared method.

AB - Longitudinal atlas construction is a challenging task in medical image analysis. Given a set of longitudinal images of different subjects, the task is how to construct the unbias longitudinal atlas sequence reflecting the anatomical changes over time. In this paper, a novel longitudinal atlas construction framework is proposed. The main contributions of the proposed method lie in the following aspects: (1) Subject-specific longitudinal information is captured by establishing a robust growth model for each subject. (2) The trajectory constraints are enforced for both subject image sequences and the atlas sequence, and only one transformation is needed for each subject to map its image sequence to the atlas sequence while preserving the temporal correspondence. (3) The longitudinal atlases are estimated by groupwise registration and kernel regression, thus no explicit template is used and the atlases are constructed without introducing bias due to the selection of the explicit template. (4) The proposed method is general, where the number of longitudinal images of each subject and the time points at which the images are taken can be different. The proposed method is evaluated on a longitudinal database and compared with a state-of-the-art longitudinal atlas construction method. Experimental results show that the proposed method achieves more consistent spatial-temporal correspondence as well as higher registration accuracy than the compared method.

UR - http://www.scopus.com/inward/record.url?scp=79959593374&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959593374&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-22092-0_24

DO - 10.1007/978-3-642-22092-0_24

M3 - Conference contribution

SN - 9783642220913

VL - 6801 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 283

EP - 295

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -