Computational analysis of LDDMM for brain mapping

Can Ceritoglu, Xiaoying Tang, Margaret Chow, Darian Hadjiabadi, Damish Shah, Timothy Brown, Muhammad H. Burhanullah, Huong Trinh, John T. Hsu, Katarina A. Ament, Deana Crocetti, Susumu Mori, Stewart H Mostofsky, Steven Yantis, Michael I. Miller, J. Tilak Ratnanather

Research output: Contribution to journalArticle

Abstract

One goal of computational anatomy (CA) is to develop tools to accurately segment brain structures in healthy and diseased subjects. In this paper, we examine the performance and complexity of such segmentation in the framework of the large deformation diffeomorphic metric mapping (LDDMM) registration method with reference to atlases and parameters. First we report the application of a multi-atlas segmentation approach to define basal ganglia structures in healthy and diseased kids' brains. The segmentation accuracy of the multi-atlas approach is compared with the single atlas LDDMM implementation and two state-of-the-art segmentation algorithms-Freesurfer and FSL-by computing the overlap errors between automatic and manual segmentations of the six basal ganglia nuclei in healthy subjects as well as subjects with diseases including ADHD and Autism. The high accuracy of multi-atlas segmentation is obtained at the cost of increasing the computational complexity because of the calculations necessary between the atlases and a subject. Second, we examine the effect of parameters on total LDDMM computation time and segmentation accuracy for basal ganglia structures. Single atlas LDDMM method is used to automatically segment the structures in a population of 16 subjects using different sets of parameters. The results show that a cascade approach and using fewer time steps can reduce computational complexity as much as five times while maintaining reliable segmentations.

Original languageEnglish (US)
Article numberArticle 151
JournalFrontiers in Neuroscience
Issue number7 AUG
DOIs
StatePublished - 2013

Fingerprint

Brain Mapping
Atlases
Basal Ganglia
Healthy Volunteers
Brain Diseases
Autistic Disorder
Anatomy
Brain

Keywords

  • Brain mapping
  • Computational anatomy
  • LDDMM
  • Subcortical segmentation

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Ceritoglu, C., Tang, X., Chow, M., Hadjiabadi, D., Shah, D., Brown, T., ... Tilak Ratnanather, J. (2013). Computational analysis of LDDMM for brain mapping. Frontiers in Neuroscience, (7 AUG), [Article 151]. https://doi.org/10.3389/fnins.2013.00151

Computational analysis of LDDMM for brain mapping. / Ceritoglu, Can; Tang, Xiaoying; Chow, Margaret; Hadjiabadi, Darian; Shah, Damish; Brown, Timothy; Burhanullah, Muhammad H.; Trinh, Huong; Hsu, John T.; Ament, Katarina A.; Crocetti, Deana; Mori, Susumu; Mostofsky, Stewart H; Yantis, Steven; Miller, Michael I.; Tilak Ratnanather, J.

In: Frontiers in Neuroscience, No. 7 AUG, Article 151, 2013.

Research output: Contribution to journalArticle

Ceritoglu, C, Tang, X, Chow, M, Hadjiabadi, D, Shah, D, Brown, T, Burhanullah, MH, Trinh, H, Hsu, JT, Ament, KA, Crocetti, D, Mori, S, Mostofsky, SH, Yantis, S, Miller, MI & Tilak Ratnanather, J 2013, 'Computational analysis of LDDMM for brain mapping', Frontiers in Neuroscience, no. 7 AUG, Article 151. https://doi.org/10.3389/fnins.2013.00151
Ceritoglu C, Tang X, Chow M, Hadjiabadi D, Shah D, Brown T et al. Computational analysis of LDDMM for brain mapping. Frontiers in Neuroscience. 2013;(7 AUG). Article 151. https://doi.org/10.3389/fnins.2013.00151
Ceritoglu, Can ; Tang, Xiaoying ; Chow, Margaret ; Hadjiabadi, Darian ; Shah, Damish ; Brown, Timothy ; Burhanullah, Muhammad H. ; Trinh, Huong ; Hsu, John T. ; Ament, Katarina A. ; Crocetti, Deana ; Mori, Susumu ; Mostofsky, Stewart H ; Yantis, Steven ; Miller, Michael I. ; Tilak Ratnanather, J. / Computational analysis of LDDMM for brain mapping. In: Frontiers in Neuroscience. 2013 ; No. 7 AUG.
@article{141f1b5acf8c46aca02ff7fd2aca9164,
title = "Computational analysis of LDDMM for brain mapping",
abstract = "One goal of computational anatomy (CA) is to develop tools to accurately segment brain structures in healthy and diseased subjects. In this paper, we examine the performance and complexity of such segmentation in the framework of the large deformation diffeomorphic metric mapping (LDDMM) registration method with reference to atlases and parameters. First we report the application of a multi-atlas segmentation approach to define basal ganglia structures in healthy and diseased kids' brains. The segmentation accuracy of the multi-atlas approach is compared with the single atlas LDDMM implementation and two state-of-the-art segmentation algorithms-Freesurfer and FSL-by computing the overlap errors between automatic and manual segmentations of the six basal ganglia nuclei in healthy subjects as well as subjects with diseases including ADHD and Autism. The high accuracy of multi-atlas segmentation is obtained at the cost of increasing the computational complexity because of the calculations necessary between the atlases and a subject. Second, we examine the effect of parameters on total LDDMM computation time and segmentation accuracy for basal ganglia structures. Single atlas LDDMM method is used to automatically segment the structures in a population of 16 subjects using different sets of parameters. The results show that a cascade approach and using fewer time steps can reduce computational complexity as much as five times while maintaining reliable segmentations.",
keywords = "Brain mapping, Computational anatomy, LDDMM, Subcortical segmentation",
author = "Can Ceritoglu and Xiaoying Tang and Margaret Chow and Darian Hadjiabadi and Damish Shah and Timothy Brown and Burhanullah, {Muhammad H.} and Huong Trinh and Hsu, {John T.} and Ament, {Katarina A.} and Deana Crocetti and Susumu Mori and Mostofsky, {Stewart H} and Steven Yantis and Miller, {Michael I.} and {Tilak Ratnanather}, J.",
year = "2013",
doi = "10.3389/fnins.2013.00151",
language = "English (US)",
journal = "Frontiers in Neuroscience",
issn = "1662-4548",
publisher = "Frontiers Research Foundation",
number = "7 AUG",

}

TY - JOUR

T1 - Computational analysis of LDDMM for brain mapping

AU - Ceritoglu, Can

AU - Tang, Xiaoying

AU - Chow, Margaret

AU - Hadjiabadi, Darian

AU - Shah, Damish

AU - Brown, Timothy

AU - Burhanullah, Muhammad H.

AU - Trinh, Huong

AU - Hsu, John T.

AU - Ament, Katarina A.

AU - Crocetti, Deana

AU - Mori, Susumu

AU - Mostofsky, Stewart H

AU - Yantis, Steven

AU - Miller, Michael I.

AU - Tilak Ratnanather, J.

PY - 2013

Y1 - 2013

N2 - One goal of computational anatomy (CA) is to develop tools to accurately segment brain structures in healthy and diseased subjects. In this paper, we examine the performance and complexity of such segmentation in the framework of the large deformation diffeomorphic metric mapping (LDDMM) registration method with reference to atlases and parameters. First we report the application of a multi-atlas segmentation approach to define basal ganglia structures in healthy and diseased kids' brains. The segmentation accuracy of the multi-atlas approach is compared with the single atlas LDDMM implementation and two state-of-the-art segmentation algorithms-Freesurfer and FSL-by computing the overlap errors between automatic and manual segmentations of the six basal ganglia nuclei in healthy subjects as well as subjects with diseases including ADHD and Autism. The high accuracy of multi-atlas segmentation is obtained at the cost of increasing the computational complexity because of the calculations necessary between the atlases and a subject. Second, we examine the effect of parameters on total LDDMM computation time and segmentation accuracy for basal ganglia structures. Single atlas LDDMM method is used to automatically segment the structures in a population of 16 subjects using different sets of parameters. The results show that a cascade approach and using fewer time steps can reduce computational complexity as much as five times while maintaining reliable segmentations.

AB - One goal of computational anatomy (CA) is to develop tools to accurately segment brain structures in healthy and diseased subjects. In this paper, we examine the performance and complexity of such segmentation in the framework of the large deformation diffeomorphic metric mapping (LDDMM) registration method with reference to atlases and parameters. First we report the application of a multi-atlas segmentation approach to define basal ganglia structures in healthy and diseased kids' brains. The segmentation accuracy of the multi-atlas approach is compared with the single atlas LDDMM implementation and two state-of-the-art segmentation algorithms-Freesurfer and FSL-by computing the overlap errors between automatic and manual segmentations of the six basal ganglia nuclei in healthy subjects as well as subjects with diseases including ADHD and Autism. The high accuracy of multi-atlas segmentation is obtained at the cost of increasing the computational complexity because of the calculations necessary between the atlases and a subject. Second, we examine the effect of parameters on total LDDMM computation time and segmentation accuracy for basal ganglia structures. Single atlas LDDMM method is used to automatically segment the structures in a population of 16 subjects using different sets of parameters. The results show that a cascade approach and using fewer time steps can reduce computational complexity as much as five times while maintaining reliable segmentations.

KW - Brain mapping

KW - Computational anatomy

KW - LDDMM

KW - Subcortical segmentation

UR - http://www.scopus.com/inward/record.url?scp=84901051064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901051064&partnerID=8YFLogxK

U2 - 10.3389/fnins.2013.00151

DO - 10.3389/fnins.2013.00151

M3 - Article

C2 - 23986653

AN - SCOPUS:84901051064

JO - Frontiers in Neuroscience

JF - Frontiers in Neuroscience

SN - 1662-4548

IS - 7 AUG

M1 - Article 151

ER -