Hierachical spherical harmonics based deformable HARDI registration

Pew Thian Yap, Yasheng Chen, Hongyu An, John H. Gilmore, Weili Lin, Dinggang Shen

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

1 Citation (Scopus)

Abstract

In contrast to the more common Diffusion Tensor Imaging (DTI), High Angular Resolution Diffusion Imaging (HARDI) allows superior delineation of angular microstructures of brain white matter, and makes possible multiple-fiber modeling of each voxel for better characterization of brain connectivity. However, in the context of image registration, the question of how much information is needed for satisfactory alignment remains unanswered. Low order representation of the diffusivity information is generally more robust than the higher order representation, but the latter gives more information for correct fiber tract alignment. However, higher order representation, when naïvely utilized, might not necessarily be conducive to improving registration accuracy since similar structures with significant orientation differences prior to proper alignment might be mistakenly taken as non-matching structures. We propose in this paper a hierarchical spherical harmonics based registration algorithm which utilizes the wealth of information provided by HARDI in a more principled means. The image volumes are first registered using robust, relatively direction invariant features derived from the diffusion-attenuation profile, and their alignment is then refined using spherical harmonic (SH) representation of gradually increasing order. This progression of SH representation from non-directional, single-directional to multi-directional representation provides a systematic means of extracting directional information from the HARDI data. Experimental results show a significant increase in registration accuracy over a state-of-the-art DTI registration algorithm.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages228-236
Number of pages9
Volume6326 LNCS
DOIs
Publication statusPublished - 2010 Nov 9
Externally publishedYes
Event5th International Workshop on Medical Imaging and Augmented Reality, MIAR 2010 - Beijing, China
Duration: 2010 Sep 192010 Sep 20

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6326 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other5th International Workshop on Medical Imaging and Augmented Reality, MIAR 2010
CountryChina
CityBeijing
Period10/9/1910/9/20

Fingerprint

High Angular Resolution
Spherical Harmonics
Registration
Imaging
Imaging techniques
Diffusion tensor imaging
Alignment
Brain
Tensor
Fiber
Higher Order
Fibers
Image registration
Voxel
Image Registration
Diffusivity
Progression
Attenuation
Microstructure
Connectivity

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Yap, P. T., Chen, Y., An, H., Gilmore, J. H., Lin, W., & Shen, D. (2010). Hierachical spherical harmonics based deformable HARDI registration. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6326 LNCS, pp. 228-236). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6326 LNCS). https://doi.org/10.1007/978-3-642-15699-1_24

Hierachical spherical harmonics based deformable HARDI registration. / Yap, Pew Thian; Chen, Yasheng; An, Hongyu; Gilmore, John H.; Lin, Weili; Shen, Dinggang.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6326 LNCS 2010. p. 228-236 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6326 LNCS).

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

Yap, PT, Chen, Y, An, H, Gilmore, JH, Lin, W & Shen, D 2010, Hierachical spherical harmonics based deformable HARDI registration. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 6326 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6326 LNCS, pp. 228-236, 5th International Workshop on Medical Imaging and Augmented Reality, MIAR 2010, Beijing, China, 10/9/19. https://doi.org/10.1007/978-3-642-15699-1_24
Yap PT, Chen Y, An H, Gilmore JH, Lin W, Shen D. Hierachical spherical harmonics based deformable HARDI registration. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6326 LNCS. 2010. p. 228-236. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-15699-1_24
Yap, Pew Thian ; Chen, Yasheng ; An, Hongyu ; Gilmore, John H. ; Lin, Weili ; Shen, Dinggang. / Hierachical spherical harmonics based deformable HARDI registration. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 6326 LNCS 2010. pp. 228-236 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c2615691337b452f9b4c6d2742a8317f,
title = "Hierachical spherical harmonics based deformable HARDI registration",
abstract = "In contrast to the more common Diffusion Tensor Imaging (DTI), High Angular Resolution Diffusion Imaging (HARDI) allows superior delineation of angular microstructures of brain white matter, and makes possible multiple-fiber modeling of each voxel for better characterization of brain connectivity. However, in the context of image registration, the question of how much information is needed for satisfactory alignment remains unanswered. Low order representation of the diffusivity information is generally more robust than the higher order representation, but the latter gives more information for correct fiber tract alignment. However, higher order representation, when na{\"i}vely utilized, might not necessarily be conducive to improving registration accuracy since similar structures with significant orientation differences prior to proper alignment might be mistakenly taken as non-matching structures. We propose in this paper a hierarchical spherical harmonics based registration algorithm which utilizes the wealth of information provided by HARDI in a more principled means. The image volumes are first registered using robust, relatively direction invariant features derived from the diffusion-attenuation profile, and their alignment is then refined using spherical harmonic (SH) representation of gradually increasing order. This progression of SH representation from non-directional, single-directional to multi-directional representation provides a systematic means of extracting directional information from the HARDI data. Experimental results show a significant increase in registration accuracy over a state-of-the-art DTI registration algorithm.",
author = "Yap, {Pew Thian} and Yasheng Chen and Hongyu An and Gilmore, {John H.} and Weili Lin and Dinggang Shen",
year = "2010",
month = "11",
day = "9",
doi = "10.1007/978-3-642-15699-1_24",
language = "English",
isbn = "3642156983",
volume = "6326 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "228--236",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Hierachical spherical harmonics based deformable HARDI registration

AU - Yap, Pew Thian

AU - Chen, Yasheng

AU - An, Hongyu

AU - Gilmore, John H.

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2010/11/9

Y1 - 2010/11/9

N2 - In contrast to the more common Diffusion Tensor Imaging (DTI), High Angular Resolution Diffusion Imaging (HARDI) allows superior delineation of angular microstructures of brain white matter, and makes possible multiple-fiber modeling of each voxel for better characterization of brain connectivity. However, in the context of image registration, the question of how much information is needed for satisfactory alignment remains unanswered. Low order representation of the diffusivity information is generally more robust than the higher order representation, but the latter gives more information for correct fiber tract alignment. However, higher order representation, when naïvely utilized, might not necessarily be conducive to improving registration accuracy since similar structures with significant orientation differences prior to proper alignment might be mistakenly taken as non-matching structures. We propose in this paper a hierarchical spherical harmonics based registration algorithm which utilizes the wealth of information provided by HARDI in a more principled means. The image volumes are first registered using robust, relatively direction invariant features derived from the diffusion-attenuation profile, and their alignment is then refined using spherical harmonic (SH) representation of gradually increasing order. This progression of SH representation from non-directional, single-directional to multi-directional representation provides a systematic means of extracting directional information from the HARDI data. Experimental results show a significant increase in registration accuracy over a state-of-the-art DTI registration algorithm.

AB - In contrast to the more common Diffusion Tensor Imaging (DTI), High Angular Resolution Diffusion Imaging (HARDI) allows superior delineation of angular microstructures of brain white matter, and makes possible multiple-fiber modeling of each voxel for better characterization of brain connectivity. However, in the context of image registration, the question of how much information is needed for satisfactory alignment remains unanswered. Low order representation of the diffusivity information is generally more robust than the higher order representation, but the latter gives more information for correct fiber tract alignment. However, higher order representation, when naïvely utilized, might not necessarily be conducive to improving registration accuracy since similar structures with significant orientation differences prior to proper alignment might be mistakenly taken as non-matching structures. We propose in this paper a hierarchical spherical harmonics based registration algorithm which utilizes the wealth of information provided by HARDI in a more principled means. The image volumes are first registered using robust, relatively direction invariant features derived from the diffusion-attenuation profile, and their alignment is then refined using spherical harmonic (SH) representation of gradually increasing order. This progression of SH representation from non-directional, single-directional to multi-directional representation provides a systematic means of extracting directional information from the HARDI data. Experimental results show a significant increase in registration accuracy over a state-of-the-art DTI registration algorithm.

UR - http://www.scopus.com/inward/record.url?scp=78049421771&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049421771&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-15699-1_24

DO - 10.1007/978-3-642-15699-1_24

M3 - Conference contribution

AN - SCOPUS:78049421771

SN - 3642156983

SN - 9783642156984

VL - 6326 LNCS

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

SP - 228

EP - 236

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

ER -