Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images: Application in multiple sclerosis

Mark A. Horsfield, Stefania Sala, Mohit Neema, Martina Absinta, Anshika Bakshi, Maria Pia Sormani, Maria A. Rocca, Rohit Bakshi, Massimo Filippi

Research output: Contribution to journalArticle

Abstract

A new semi-automatic method for segmenting the spinal cord from MR images is presented. The method is based on an active surface (AS) model of the cord surface, with intrinsic smoothness constraints. The model is initialized by the user marking the approximate cord center-line on a few representative slices, and the compact surface parametrization results in a rapid segmentation, taking on the order of 1 min. Using 3-D acquired T1-weighted images of the cervical spine from human controls and patients with multiple sclerosis, the intra- and inter-observer reproducibilities were evaluated, and compared favorably with an existing cord segmentation method. While the AS method overestimated the cord area by approximately 14% compared to manual outlining, correlations between cord cross-sectional area and clinical disability scores confirmed the relevance of the new method in measuring cord atrophy in multiple sclerosis. Segmentation of the cord from 2-D multi-slice T2-weighted images is also demonstrated over the cervical and thoracic region. Since the cord center-line is an intrinsic parameter extracted as part of the segmentation process, the image can be resampled such that the center-line forms one coordinate axis of a new image, allowing simple visualization of the cord structure and pathology; this could find wider application in standard radiological practice.

Original languageEnglish (US)
Pages (from-to)446-455
Number of pages10
JournalNeuroImage
Volume50
Issue number2
DOIs
StatePublished - Jan 1 2010
Externally publishedYes

Fingerprint

Multiple Sclerosis
Spinal Cord
Magnetic Resonance Spectroscopy
Atrophy
Spine
Thorax
Pathology

ASJC Scopus subject areas

  • Neurology
  • Cognitive Neuroscience

Cite this

Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images : Application in multiple sclerosis. / Horsfield, Mark A.; Sala, Stefania; Neema, Mohit; Absinta, Martina; Bakshi, Anshika; Sormani, Maria Pia; Rocca, Maria A.; Bakshi, Rohit; Filippi, Massimo.

In: NeuroImage, Vol. 50, No. 2, 01.01.2010, p. 446-455.

Research output: Contribution to journalArticle

Horsfield, MA, Sala, S, Neema, M, Absinta, M, Bakshi, A, Sormani, MP, Rocca, MA, Bakshi, R & Filippi, M 2010, 'Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images: Application in multiple sclerosis', NeuroImage, vol. 50, no. 2, pp. 446-455. https://doi.org/10.1016/j.neuroimage.2009.12.121
Horsfield, Mark A. ; Sala, Stefania ; Neema, Mohit ; Absinta, Martina ; Bakshi, Anshika ; Sormani, Maria Pia ; Rocca, Maria A. ; Bakshi, Rohit ; Filippi, Massimo. / Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images : Application in multiple sclerosis. In: NeuroImage. 2010 ; Vol. 50, No. 2. pp. 446-455.
@article{8b0de59d5fae4b32a6b78e1cd39883fd,
title = "Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images: Application in multiple sclerosis",
abstract = "A new semi-automatic method for segmenting the spinal cord from MR images is presented. The method is based on an active surface (AS) model of the cord surface, with intrinsic smoothness constraints. The model is initialized by the user marking the approximate cord center-line on a few representative slices, and the compact surface parametrization results in a rapid segmentation, taking on the order of 1 min. Using 3-D acquired T1-weighted images of the cervical spine from human controls and patients with multiple sclerosis, the intra- and inter-observer reproducibilities were evaluated, and compared favorably with an existing cord segmentation method. While the AS method overestimated the cord area by approximately 14{\%} compared to manual outlining, correlations between cord cross-sectional area and clinical disability scores confirmed the relevance of the new method in measuring cord atrophy in multiple sclerosis. Segmentation of the cord from 2-D multi-slice T2-weighted images is also demonstrated over the cervical and thoracic region. Since the cord center-line is an intrinsic parameter extracted as part of the segmentation process, the image can be resampled such that the center-line forms one coordinate axis of a new image, allowing simple visualization of the cord structure and pathology; this could find wider application in standard radiological practice.",
author = "Horsfield, {Mark A.} and Stefania Sala and Mohit Neema and Martina Absinta and Anshika Bakshi and Sormani, {Maria Pia} and Rocca, {Maria A.} and Rohit Bakshi and Massimo Filippi",
year = "2010",
month = "1",
day = "1",
doi = "10.1016/j.neuroimage.2009.12.121",
language = "English (US)",
volume = "50",
pages = "446--455",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Rapid semi-automatic segmentation of the spinal cord from magnetic resonance images

T2 - Application in multiple sclerosis

AU - Horsfield, Mark A.

AU - Sala, Stefania

AU - Neema, Mohit

AU - Absinta, Martina

AU - Bakshi, Anshika

AU - Sormani, Maria Pia

AU - Rocca, Maria A.

AU - Bakshi, Rohit

AU - Filippi, Massimo

PY - 2010/1/1

Y1 - 2010/1/1

N2 - A new semi-automatic method for segmenting the spinal cord from MR images is presented. The method is based on an active surface (AS) model of the cord surface, with intrinsic smoothness constraints. The model is initialized by the user marking the approximate cord center-line on a few representative slices, and the compact surface parametrization results in a rapid segmentation, taking on the order of 1 min. Using 3-D acquired T1-weighted images of the cervical spine from human controls and patients with multiple sclerosis, the intra- and inter-observer reproducibilities were evaluated, and compared favorably with an existing cord segmentation method. While the AS method overestimated the cord area by approximately 14% compared to manual outlining, correlations between cord cross-sectional area and clinical disability scores confirmed the relevance of the new method in measuring cord atrophy in multiple sclerosis. Segmentation of the cord from 2-D multi-slice T2-weighted images is also demonstrated over the cervical and thoracic region. Since the cord center-line is an intrinsic parameter extracted as part of the segmentation process, the image can be resampled such that the center-line forms one coordinate axis of a new image, allowing simple visualization of the cord structure and pathology; this could find wider application in standard radiological practice.

AB - A new semi-automatic method for segmenting the spinal cord from MR images is presented. The method is based on an active surface (AS) model of the cord surface, with intrinsic smoothness constraints. The model is initialized by the user marking the approximate cord center-line on a few representative slices, and the compact surface parametrization results in a rapid segmentation, taking on the order of 1 min. Using 3-D acquired T1-weighted images of the cervical spine from human controls and patients with multiple sclerosis, the intra- and inter-observer reproducibilities were evaluated, and compared favorably with an existing cord segmentation method. While the AS method overestimated the cord area by approximately 14% compared to manual outlining, correlations between cord cross-sectional area and clinical disability scores confirmed the relevance of the new method in measuring cord atrophy in multiple sclerosis. Segmentation of the cord from 2-D multi-slice T2-weighted images is also demonstrated over the cervical and thoracic region. Since the cord center-line is an intrinsic parameter extracted as part of the segmentation process, the image can be resampled such that the center-line forms one coordinate axis of a new image, allowing simple visualization of the cord structure and pathology; this could find wider application in standard radiological practice.

UR - http://www.scopus.com/inward/record.url?scp=77952294568&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952294568&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2009.12.121

DO - 10.1016/j.neuroimage.2009.12.121

M3 - Article

C2 - 20060481

AN - SCOPUS:77952294568

VL - 50

SP - 446

EP - 455

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 2

ER -