Limits...
Spinal Cord Segmentation by One Dimensional Normalized Template Matching: A Novel, Quantitative Technique to Analyze Advanced Magnetic Resonance Imaging Data.

Cadotte A, Cadotte DW, Livne M, Cohen-Adad J, Fleet D, Mikulis D, Fehlings MG - PLoS ONE (2015)

Bottom Line: Semi-automatic validation was performed by comparing the two manual raters to the semi-automatic method (n = 10).The absolute variation in cross-sectional area was comparable for the semi-automatic method versus manual segmentation when compared to inter-rater manual segmentation.The results demonstrate that this novel segmentation method performs as well as a manual rater for most segmentation metrics.

View Article: PubMed Central - PubMed

Affiliation: Department of Surgery, Division of Neurosurgery, University of Toronto, Toronto, Ontario, Canada.

ABSTRACT
Spinal cord segmentation is a developing area of research intended to aid the processing and interpretation of advanced magnetic resonance imaging (MRI). For example, high resolution three-dimensional volumes can be segmented to provide a measurement of spinal cord atrophy. Spinal cord segmentation is difficult due to the variety of MRI contrasts and the variation in human anatomy. In this study we propose a new method of spinal cord segmentation based on one-dimensional template matching and provide several metrics that can be used to compare with other segmentation methods. A set of ground-truth data from 10 subjects was manually-segmented by two different raters. These ground truth data formed the basis of the segmentation algorithm. A user was required to manually initialize the spinal cord center-line on new images, taking less than one minute. Template matching was used to segment the new cord and a refined center line was calculated based on multiple centroids within the segmentation. Arc distances down the spinal cord and cross-sectional areas were calculated. Inter-rater validation was performed by comparing two manual raters (n = 10). Semi-automatic validation was performed by comparing the two manual raters to the semi-automatic method (n = 10). Comparing the semi-automatic method to one of the raters yielded a Dice coefficient of 0.91 +/- 0.02 for ten subjects, a mean distance between spinal cord center lines of 0.32 +/- 0.08 mm, and a Hausdorff distance of 1.82 +/- 0.33 mm. The absolute variation in cross-sectional area was comparable for the semi-automatic method versus manual segmentation when compared to inter-rater manual segmentation. The results demonstrate that this novel segmentation method performs as well as a manual rater for most segmentation metrics. It offers a new approach to study spinal cord disease and to quantitatively track changes within the spinal cord in an individual case and across cohorts of subjects.

No MeSH data available.


Related in: MedlinePlus

An axial slice of the MR image (left) and the gradient image (right) of the cervical spinal cord.πThe light green shading illustrates the manual segmentation. The red radial line rotating about the spinal cord centerline axis was used for extracting the MRI signal.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4596853&req=5

pone.0139323.g002: An axial slice of the MR image (left) and the gradient image (right) of the cervical spinal cord.πThe light green shading illustrates the manual segmentation. The red radial line rotating about the spinal cord centerline axis was used for extracting the MRI signal.

Mentions: To create the ground truth data, one-dimensional template arrays are created from the set of radial lines emanating from the center of the spinal cord of the manually segmented image. The center of the manually segmented spinal cord was calculated by taking the centroid of the segmentation in all axial planes along the length of the cord. Radial lines emanate from the center point in two degree increments in the axial plane. The nearest voxels to the calculated radial line (using 0.1mm increments along the radial line) were recorded. 70 voxels were used to create each radial line. The first image shown in Fig 2, an axial slice of a MR image, shows the segmented spinal cord (shaded light green) and a single radial line emanating from the center of the spinal cord to 70 voxels away from the center of the spinal cord. The second image in Fig 2 shows the magnitude of the gradient of the image (the “gradient image”), for the same axial slice. The gradient image shows areas of contrast, which are more pronounced at the intersection of the CSF and spinal cord.


Spinal Cord Segmentation by One Dimensional Normalized Template Matching: A Novel, Quantitative Technique to Analyze Advanced Magnetic Resonance Imaging Data.

Cadotte A, Cadotte DW, Livne M, Cohen-Adad J, Fleet D, Mikulis D, Fehlings MG - PLoS ONE (2015)

An axial slice of the MR image (left) and the gradient image (right) of the cervical spinal cord.πThe light green shading illustrates the manual segmentation. The red radial line rotating about the spinal cord centerline axis was used for extracting the MRI signal.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4596853&req=5

pone.0139323.g002: An axial slice of the MR image (left) and the gradient image (right) of the cervical spinal cord.πThe light green shading illustrates the manual segmentation. The red radial line rotating about the spinal cord centerline axis was used for extracting the MRI signal.
Mentions: To create the ground truth data, one-dimensional template arrays are created from the set of radial lines emanating from the center of the spinal cord of the manually segmented image. The center of the manually segmented spinal cord was calculated by taking the centroid of the segmentation in all axial planes along the length of the cord. Radial lines emanate from the center point in two degree increments in the axial plane. The nearest voxels to the calculated radial line (using 0.1mm increments along the radial line) were recorded. 70 voxels were used to create each radial line. The first image shown in Fig 2, an axial slice of a MR image, shows the segmented spinal cord (shaded light green) and a single radial line emanating from the center of the spinal cord to 70 voxels away from the center of the spinal cord. The second image in Fig 2 shows the magnitude of the gradient of the image (the “gradient image”), for the same axial slice. The gradient image shows areas of contrast, which are more pronounced at the intersection of the CSF and spinal cord.

Bottom Line: Semi-automatic validation was performed by comparing the two manual raters to the semi-automatic method (n = 10).The absolute variation in cross-sectional area was comparable for the semi-automatic method versus manual segmentation when compared to inter-rater manual segmentation.The results demonstrate that this novel segmentation method performs as well as a manual rater for most segmentation metrics.

View Article: PubMed Central - PubMed

Affiliation: Department of Surgery, Division of Neurosurgery, University of Toronto, Toronto, Ontario, Canada.

ABSTRACT
Spinal cord segmentation is a developing area of research intended to aid the processing and interpretation of advanced magnetic resonance imaging (MRI). For example, high resolution three-dimensional volumes can be segmented to provide a measurement of spinal cord atrophy. Spinal cord segmentation is difficult due to the variety of MRI contrasts and the variation in human anatomy. In this study we propose a new method of spinal cord segmentation based on one-dimensional template matching and provide several metrics that can be used to compare with other segmentation methods. A set of ground-truth data from 10 subjects was manually-segmented by two different raters. These ground truth data formed the basis of the segmentation algorithm. A user was required to manually initialize the spinal cord center-line on new images, taking less than one minute. Template matching was used to segment the new cord and a refined center line was calculated based on multiple centroids within the segmentation. Arc distances down the spinal cord and cross-sectional areas were calculated. Inter-rater validation was performed by comparing two manual raters (n = 10). Semi-automatic validation was performed by comparing the two manual raters to the semi-automatic method (n = 10). Comparing the semi-automatic method to one of the raters yielded a Dice coefficient of 0.91 +/- 0.02 for ten subjects, a mean distance between spinal cord center lines of 0.32 +/- 0.08 mm, and a Hausdorff distance of 1.82 +/- 0.33 mm. The absolute variation in cross-sectional area was comparable for the semi-automatic method versus manual segmentation when compared to inter-rater manual segmentation. The results demonstrate that this novel segmentation method performs as well as a manual rater for most segmentation metrics. It offers a new approach to study spinal cord disease and to quantitatively track changes within the spinal cord in an individual case and across cohorts of subjects.

No MeSH data available.


Related in: MedlinePlus