Automatic and reliable segmentation of spinal canals in low-resolution, low-contrast CT images

Qian Wang, Le Lu, Dijia Wu, Noha El-Zehiry, Dinggang Shen, Kevin S. Zhou

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Accurate segmentation of spinal canals in Computed Tomography (CT) images is an important task in many related studies. In this paper, we propose an automatic segmentation method and apply it to our highly challenging 110 datasets from the CT channel of PET-CT scans.We adapt the interactive random-walks (RW) segmentation algorithm to be fully automatic which is initialized with robust voxelwise classification using Haar features and probabilistic boosting tree. One-shot RW is able to estimate yet imperfect segmentation. We then refine the topology of the segmented spinal canal leading to improved seeds or boundary conditions of RW. Therefore, by iteratively optimizing the spinal canal topology and running RW segmentation, satisfactory segmentation results can be acquired within only a few iterations. Our experiments validate the capability of the proposed method with promising segmentation performance, even though the resolution and the contrast of our datasets are low.

Original languageEnglish
Pages (from-to)15-24
Number of pages10
JournalLecture Notes in Computational Vision and Biomechanics
Volume17
DOIs
Publication statusPublished - 2014
Externally publishedYes

Fingerprint

Canals
Tomography
Topology
Seed
Boundary conditions
Experiments

ASJC Scopus subject areas

  • Signal Processing
  • Biomedical Engineering
  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Mechanical Engineering

Cite this

Automatic and reliable segmentation of spinal canals in low-resolution, low-contrast CT images. / Wang, Qian; Lu, Le; Wu, Dijia; El-Zehiry, Noha; Shen, Dinggang; Zhou, Kevin S.

In: Lecture Notes in Computational Vision and Biomechanics, Vol. 17, 2014, p. 15-24.

Research output: Contribution to journalArticle

@article{d0d188532c484b6696c107f0ba5d4ddf,
title = "Automatic and reliable segmentation of spinal canals in low-resolution, low-contrast CT images",
abstract = "Accurate segmentation of spinal canals in Computed Tomography (CT) images is an important task in many related studies. In this paper, we propose an automatic segmentation method and apply it to our highly challenging 110 datasets from the CT channel of PET-CT scans.We adapt the interactive random-walks (RW) segmentation algorithm to be fully automatic which is initialized with robust voxelwise classification using Haar features and probabilistic boosting tree. One-shot RW is able to estimate yet imperfect segmentation. We then refine the topology of the segmented spinal canal leading to improved seeds or boundary conditions of RW. Therefore, by iteratively optimizing the spinal canal topology and running RW segmentation, satisfactory segmentation results can be acquired within only a few iterations. Our experiments validate the capability of the proposed method with promising segmentation performance, even though the resolution and the contrast of our datasets are low.",
author = "Qian Wang and Le Lu and Dijia Wu and Noha El-Zehiry and Dinggang Shen and Zhou, {Kevin S.}",
year = "2014",
doi = "10.1007/978-3-319-07269-2_2",
language = "English",
volume = "17",
pages = "15--24",
journal = "Lecture Notes in Computational Vision and Biomechanics",
issn = "2212-9413",
publisher = "Springer International Publishing AG",

}

TY - JOUR

T1 - Automatic and reliable segmentation of spinal canals in low-resolution, low-contrast CT images

AU - Wang, Qian

AU - Lu, Le

AU - Wu, Dijia

AU - El-Zehiry, Noha

AU - Shen, Dinggang

AU - Zhou, Kevin S.

PY - 2014

Y1 - 2014

N2 - Accurate segmentation of spinal canals in Computed Tomography (CT) images is an important task in many related studies. In this paper, we propose an automatic segmentation method and apply it to our highly challenging 110 datasets from the CT channel of PET-CT scans.We adapt the interactive random-walks (RW) segmentation algorithm to be fully automatic which is initialized with robust voxelwise classification using Haar features and probabilistic boosting tree. One-shot RW is able to estimate yet imperfect segmentation. We then refine the topology of the segmented spinal canal leading to improved seeds or boundary conditions of RW. Therefore, by iteratively optimizing the spinal canal topology and running RW segmentation, satisfactory segmentation results can be acquired within only a few iterations. Our experiments validate the capability of the proposed method with promising segmentation performance, even though the resolution and the contrast of our datasets are low.

AB - Accurate segmentation of spinal canals in Computed Tomography (CT) images is an important task in many related studies. In this paper, we propose an automatic segmentation method and apply it to our highly challenging 110 datasets from the CT channel of PET-CT scans.We adapt the interactive random-walks (RW) segmentation algorithm to be fully automatic which is initialized with robust voxelwise classification using Haar features and probabilistic boosting tree. One-shot RW is able to estimate yet imperfect segmentation. We then refine the topology of the segmented spinal canal leading to improved seeds or boundary conditions of RW. Therefore, by iteratively optimizing the spinal canal topology and running RW segmentation, satisfactory segmentation results can be acquired within only a few iterations. Our experiments validate the capability of the proposed method with promising segmentation performance, even though the resolution and the contrast of our datasets are low.

UR - http://www.scopus.com/inward/record.url?scp=84927587344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84927587344&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-07269-2_2

DO - 10.1007/978-3-319-07269-2_2

M3 - Article

AN - SCOPUS:84927587344

VL - 17

SP - 15

EP - 24

JO - Lecture Notes in Computational Vision and Biomechanics

JF - Lecture Notes in Computational Vision and Biomechanics

SN - 2212-9413

ER -