Diffusion specific segmentation: Skull stripping with diffusion mri data alone

The Alzheimer's Disease Neuroimaging Initiative

doi:10.1007/978-3-319-73839-0_5

Diffusion specific segmentation: Skull stripping with diffusion mri data alone

The Alzheimer's Disease Neuroimaging Initiative

Radiology

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

3 Scopus citations

Abstract

Most processing pipelines for diffusion MRI (dMRI) require an intracranial mask image to exclude voxels outside the skull, and some dMRI analyses also need a segmentation between the voxels that are primarily tissue or cerebrospinal fluid (CSF). dMRI is challenging for most segmentation methods because it usually has relatively severe image artifacts and coarse resolution. However, it does provide information about the physical properties of the material(s) in each voxel, which can be directly applied to segmentation. We describe the training of a random forest classifier to segment dMRI into intracranial, brain, and CSF masks, and compare its results to three other segmentation methods commonly used in dMRI processing. The effect of correcting smooth spatial intensity variations on dMRI segmentation is also tested.

Original language	English (US)
Title of host publication	Computational Diffusion MRI - MICCAI Workshop, 2017
Editors	Enrico Kaden, Francesco Grussu, Lipeng Ning, Chantal M.W. Tax, Jelle Veraart
Publisher	Springer
Pages	67-80
Number of pages	14
ISBN (Print)	9783319738383
DOIs	https://doi.org/10.1007/978-3-319-73839-0_5
State	Published - 2018
Event	MICCAI Workshop on Computational Diffusion MRI, CDMRI 2017 - Quebec, Canada Duration: Sep 10 2017 → Sep 10 2017

Publication series

Name	Mathematics and Visualization
ISSN (Print)	1612-3786
ISSN (Electronic)	2197-666X

Conference

Conference	MICCAI Workshop on Computational Diffusion MRI, CDMRI 2017
Country/Territory	Canada
City	Quebec
Period	9/10/17 → 9/10/17

ASJC Scopus subject areas

Modeling and Simulation
Geometry and Topology
Computer Graphics and Computer-Aided Design
Applied Mathematics

Access to Document

10.1007/978-3-319-73839-0_5

Cite this

Diffusion specific segmentation: Skull stripping with diffusion mri data alone. / The Alzheimer's Disease Neuroimaging Initiative.
Computational Diffusion MRI - MICCAI Workshop, 2017. ed. / Enrico Kaden; Francesco Grussu; Lipeng Ning; Chantal M.W. Tax; Jelle Veraart. Springer, 2018. p. 67-80 (Mathematics and Visualization).

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

The Alzheimer's Disease Neuroimaging Initiative 2018, Diffusion specific segmentation: Skull stripping with diffusion mri data alone. in E Kaden, F Grussu, L Ning, CMW Tax & J Veraart (eds), Computational Diffusion MRI - MICCAI Workshop, 2017. Mathematics and Visualization, Springer, pp. 67-80, MICCAI Workshop on Computational Diffusion MRI, CDMRI 2017, Quebec, Canada, 9/10/17. https://doi.org/10.1007/978-3-319-73839-0_5

@inproceedings{797f0cc1be1942f5a948a82e1cd4ce00,

title = "Diffusion specific segmentation: Skull stripping with diffusion mri data alone",

abstract = "Most processing pipelines for diffusion MRI (dMRI) require an intracranial mask image to exclude voxels outside the skull, and some dMRI analyses also need a segmentation between the voxels that are primarily tissue or cerebrospinal fluid (CSF). dMRI is challenging for most segmentation methods because it usually has relatively severe image artifacts and coarse resolution. However, it does provide information about the physical properties of the material(s) in each voxel, which can be directly applied to segmentation. We describe the training of a random forest classifier to segment dMRI into intracranial, brain, and CSF masks, and compare its results to three other segmentation methods commonly used in dMRI processing. The effect of correcting smooth spatial intensity variations on dMRI segmentation is also tested.",

author = "{The Alzheimer's Disease Neuroimaging Initiative} and Reid, {Robert I.} and Zuzana Nedelska and Schwarz, {Christopher G.} and Chadwick Ward and Jack, {Clifford R.}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG, part of Springer Nature 2018.; MICCAI Workshop on Computational Diffusion MRI, CDMRI 2017 ; Conference date: 10-09-2017 Through 10-09-2017",

year = "2018",

doi = "10.1007/978-3-319-73839-0_5",

language = "English (US)",

isbn = "9783319738383",

series = "Mathematics and Visualization",

publisher = "Springer",

pages = "67--80",

editor = "Enrico Kaden and Francesco Grussu and Lipeng Ning and Tax, {Chantal M.W.} and Jelle Veraart",

booktitle = "Computational Diffusion MRI - MICCAI Workshop, 2017",

}

TY - GEN

T1 - Diffusion specific segmentation

T2 - MICCAI Workshop on Computational Diffusion MRI, CDMRI 2017

AU - The Alzheimer's Disease Neuroimaging Initiative

AU - Reid, Robert I.

AU - Nedelska, Zuzana

AU - Schwarz, Christopher G.

AU - Ward, Chadwick

AU - Jack, Clifford R.

N1 - Publisher Copyright: © Springer International Publishing AG, part of Springer Nature 2018.

PY - 2018

Y1 - 2018

N2 - Most processing pipelines for diffusion MRI (dMRI) require an intracranial mask image to exclude voxels outside the skull, and some dMRI analyses also need a segmentation between the voxels that are primarily tissue or cerebrospinal fluid (CSF). dMRI is challenging for most segmentation methods because it usually has relatively severe image artifacts and coarse resolution. However, it does provide information about the physical properties of the material(s) in each voxel, which can be directly applied to segmentation. We describe the training of a random forest classifier to segment dMRI into intracranial, brain, and CSF masks, and compare its results to three other segmentation methods commonly used in dMRI processing. The effect of correcting smooth spatial intensity variations on dMRI segmentation is also tested.

AB - Most processing pipelines for diffusion MRI (dMRI) require an intracranial mask image to exclude voxels outside the skull, and some dMRI analyses also need a segmentation between the voxels that are primarily tissue or cerebrospinal fluid (CSF). dMRI is challenging for most segmentation methods because it usually has relatively severe image artifacts and coarse resolution. However, it does provide information about the physical properties of the material(s) in each voxel, which can be directly applied to segmentation. We describe the training of a random forest classifier to segment dMRI into intracranial, brain, and CSF masks, and compare its results to three other segmentation methods commonly used in dMRI processing. The effect of correcting smooth spatial intensity variations on dMRI segmentation is also tested.

UR - http://www.scopus.com/inward/record.url?scp=85064507957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064507957&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-73839-0_5

DO - 10.1007/978-3-319-73839-0_5

M3 - Conference contribution

AN - SCOPUS:85064507957

SN - 9783319738383

T3 - Mathematics and Visualization

SP - 67

EP - 80

BT - Computational Diffusion MRI - MICCAI Workshop, 2017

A2 - Kaden, Enrico

A2 - Grussu, Francesco

A2 - Ning, Lipeng

A2 - Tax, Chantal M.W.

A2 - Veraart, Jelle

PB - Springer

Y2 - 10 September 2017 through 10 September 2017

ER -

Diffusion specific segmentation: Skull stripping with diffusion mri data alone

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this