Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI

P. Aljabar; R. Wolz; L. Srinivasan; S. Counsell; J. P. Boardman; M. Murgasova; V. Doria; M. A. Rutherford; A. D. Edwards; J. V. Hajnal; D. Rueckert

Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI

P. Aljabar^*, R. Wolz, L. Srinivasan, S. Counsell, J. P. Boardman, M. Murgasova, V. Doria, M. A. Rutherford, A. D. Edwards, J. V. Hajnal, D. Rueckert

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

MR, image data can provide many features or measures although any single measure is unlikely to comprehensively characterize the underlying morphology. We present a framework in which multiple measures are used in manifold learning steps to generate coordinate embeddings which are then combined to give an improved single representation of the population. An application to neonatal brain MRI data shows that the use of shape and appearance measures in particular leads to biologically plausible and consistent representations correlating well with clinical data. Orthogonality among the correlations suggests the embedding components relate to comparatively independent morphological features. The rapid changes that occur in brain shape and in MR image appearance during neonatal brain development justify the use of shape measures (obtained from a deformation metric) and appearance measures (obtained from image similarity). The benefit; of combining separate embeddings is demonstrated by improved correlations with clinical data and we illustrate the potential of the proposed framework in characterizing trajectories of brain development.

Original language	English
Title of host publication	MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III
Editors	T Jiang, N Navab, JPW Pluim, MA Viergever
Publisher	Springer
Pages	1-8
Number of pages	8
ISBN (Print)	978-3-642-15710-3
Publication status	Published - 2010
Event	13th International Conference on Medical Image Computing and Computer-Assisted Intervention - Beijing, United Kingdom Duration: 20 Sept 2010 → 24 Sept 2010

Publication series

Name	Lecture Notes in Computer Science
Publisher	SPRINGER-VERLAG BERLIN
Volume	6363
ISSN (Print)	0302-9743

Conference

Conference	13th International Conference on Medical Image Computing and Computer-Assisted Intervention
Country/Territory	United Kingdom
Period	20/09/10 → 24/09/10

Keywords / Materials (for Non-textual outputs)

DIMENSIONALITY REDUCTION
IMAGES
REGISTRATION
MORPHOMETRY

Cite this

Aljabar, P., Wolz, R., Srinivasan, L., Counsell, S., Boardman, J. P., Murgasova, M., Doria, V., Rutherford, M. A., Edwards, A. D., Hajnal, J. V., & Rueckert, D. (2010). Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI. In T. Jiang, N. Navab, JPW. Pluim, & MA. Viergever (Eds.), MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III (pp. 1-8). (Lecture Notes in Computer Science; Vol. 6363). Springer.

@inproceedings{01c8e304a2ab495ab59a8eb98bfd6f05,

title = "Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI",

abstract = "MR, image data can provide many features or measures although any single measure is unlikely to comprehensively characterize the underlying morphology. We present a framework in which multiple measures are used in manifold learning steps to generate coordinate embeddings which are then combined to give an improved single representation of the population. An application to neonatal brain MRI data shows that the use of shape and appearance measures in particular leads to biologically plausible and consistent representations correlating well with clinical data. Orthogonality among the correlations suggests the embedding components relate to comparatively independent morphological features. The rapid changes that occur in brain shape and in MR image appearance during neonatal brain development justify the use of shape measures (obtained from a deformation metric) and appearance measures (obtained from image similarity). The benefit; of combining separate embeddings is demonstrated by improved correlations with clinical data and we illustrate the potential of the proposed framework in characterizing trajectories of brain development.",

keywords = "DIMENSIONALITY REDUCTION, IMAGES, REGISTRATION, MORPHOMETRY",

author = "P. Aljabar and R. Wolz and L. Srinivasan and S. Counsell and Boardman, {J. P.} and M. Murgasova and V. Doria and Rutherford, {M. A.} and Edwards, {A. D.} and Hajnal, {J. V.} and D. Rueckert",

year = "2010",

language = "English",

isbn = "978-3-642-15710-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "1--8",

editor = "T Jiang and N Navab and JPW Pluim and MA Viergever",

booktitle = "MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III",

address = "United Kingdom",

note = "13th International Conference on Medical Image Computing and Computer-Assisted Intervention ; Conference date: 20-09-2010 Through 24-09-2010",

}

Aljabar, P, Wolz, R, Srinivasan, L, Counsell, S, Boardman, JP, Murgasova, M, Doria, V, Rutherford, MA, Edwards, AD, Hajnal, JV & Rueckert, D 2010, Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI. in T Jiang, N Navab, JPW Pluim & MA Viergever (eds), MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III. Lecture Notes in Computer Science, vol. 6363, Springer, pp. 1-8, 13th International Conference on Medical Image Computing and Computer-Assisted Intervention, United Kingdom, 20/09/10.

Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI. / Aljabar, P.; Wolz, R.; Srinivasan, L. et al.
MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III. ed. / T Jiang; N Navab; JPW Pluim; MA Viergever. Springer, 2010. p. 1-8 (Lecture Notes in Computer Science; Vol. 6363).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Combining Morphological Information in a Manifold Learning Framework

T2 - 13th International Conference on Medical Image Computing and Computer-Assisted Intervention

AU - Aljabar, P.

AU - Wolz, R.

AU - Srinivasan, L.

AU - Counsell, S.

AU - Boardman, J. P.

AU - Murgasova, M.

AU - Doria, V.

AU - Rutherford, M. A.

AU - Edwards, A. D.

AU - Hajnal, J. V.

AU - Rueckert, D.

PY - 2010

Y1 - 2010

N2 - MR, image data can provide many features or measures although any single measure is unlikely to comprehensively characterize the underlying morphology. We present a framework in which multiple measures are used in manifold learning steps to generate coordinate embeddings which are then combined to give an improved single representation of the population. An application to neonatal brain MRI data shows that the use of shape and appearance measures in particular leads to biologically plausible and consistent representations correlating well with clinical data. Orthogonality among the correlations suggests the embedding components relate to comparatively independent morphological features. The rapid changes that occur in brain shape and in MR image appearance during neonatal brain development justify the use of shape measures (obtained from a deformation metric) and appearance measures (obtained from image similarity). The benefit; of combining separate embeddings is demonstrated by improved correlations with clinical data and we illustrate the potential of the proposed framework in characterizing trajectories of brain development.

AB - MR, image data can provide many features or measures although any single measure is unlikely to comprehensively characterize the underlying morphology. We present a framework in which multiple measures are used in manifold learning steps to generate coordinate embeddings which are then combined to give an improved single representation of the population. An application to neonatal brain MRI data shows that the use of shape and appearance measures in particular leads to biologically plausible and consistent representations correlating well with clinical data. Orthogonality among the correlations suggests the embedding components relate to comparatively independent morphological features. The rapid changes that occur in brain shape and in MR image appearance during neonatal brain development justify the use of shape measures (obtained from a deformation metric) and appearance measures (obtained from image similarity). The benefit; of combining separate embeddings is demonstrated by improved correlations with clinical data and we illustrate the potential of the proposed framework in characterizing trajectories of brain development.

KW - DIMENSIONALITY REDUCTION

KW - IMAGES

KW - REGISTRATION

KW - MORPHOMETRY

M3 - Conference contribution

SN - 978-3-642-15710-3

T3 - Lecture Notes in Computer Science

SP - 1

EP - 8

BT - MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III

A2 - Jiang, T

A2 - Navab, N

A2 - Pluim, JPW

A2 - Viergever, MA

PB - Springer

Y2 - 20 September 2010 through 24 September 2010

ER -

Aljabar P, Wolz R, Srinivasan L, Counsell S, Boardman JP, Murgasova M et al. Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI. In Jiang T, Navab N, Pluim JPW, Viergever MA, editors, MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2010, PT III. Springer. 2010. p. 1-8. (Lecture Notes in Computer Science).

Combining Morphological Information in a Manifold Learning Framework: Application to Neonatal MRI

Abstract

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Fingerprint

Cite this