Automated segmentation of 3D US prostate images using statistical texture-based matching method

Yiqiang Zhan, Dinggang Shen

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

A novel statistical shape model is presented for automatic and accurate segmentation of prostate boundary from 3D ultrasound (US) images, using a hierarchical texture-based matching method. This method uses three steps. First, Gabor filter banks are used to capture rotation-invariant texture features at different scales and orientations. Second, different levels of texture features are integrated by a kernel support vector machine (KSVM) to optimally differentiate the prostate from surrounding tissues. Third, a statistical shape model is hierarchically deformed to the prostate boundary by robust texture and shape matching. Experimental results test the performance of the proposed method in segmenting 3D US prostate images.

Original languageEnglish
Pages (from-to)688-696
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2878
Publication statusPublished - 2003 Dec 1
Externally publishedYes

Fingerprint

Ultrasound
Texture
Prostate
Segmentation
Texture Feature
Textures
Ultrasonics
Statistical Models
Shape Matching
Gabor Filter
Rotation Invariant
Ultrasound Image
Filter Banks
3D Image
Differentiate
Gabor filters
Support Vector Machine
Filter banks
kernel
Support vector machines

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science
  • Computer Science (miscellaneous)
  • Engineering(all)

Cite this

@article{2a6d9c82e3ff48358cd92100510643e5,
title = "Automated segmentation of 3D US prostate images using statistical texture-based matching method",
abstract = "A novel statistical shape model is presented for automatic and accurate segmentation of prostate boundary from 3D ultrasound (US) images, using a hierarchical texture-based matching method. This method uses three steps. First, Gabor filter banks are used to capture rotation-invariant texture features at different scales and orientations. Second, different levels of texture features are integrated by a kernel support vector machine (KSVM) to optimally differentiate the prostate from surrounding tissues. Third, a statistical shape model is hierarchically deformed to the prostate boundary by robust texture and shape matching. Experimental results test the performance of the proposed method in segmenting 3D US prostate images.",
author = "Yiqiang Zhan and Dinggang Shen",
year = "2003",
month = "12",
day = "1",
language = "English",
volume = "2878",
pages = "688--696",
journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Automated segmentation of 3D US prostate images using statistical texture-based matching method

AU - Zhan, Yiqiang

AU - Shen, Dinggang

PY - 2003/12/1

Y1 - 2003/12/1

N2 - A novel statistical shape model is presented for automatic and accurate segmentation of prostate boundary from 3D ultrasound (US) images, using a hierarchical texture-based matching method. This method uses three steps. First, Gabor filter banks are used to capture rotation-invariant texture features at different scales and orientations. Second, different levels of texture features are integrated by a kernel support vector machine (KSVM) to optimally differentiate the prostate from surrounding tissues. Third, a statistical shape model is hierarchically deformed to the prostate boundary by robust texture and shape matching. Experimental results test the performance of the proposed method in segmenting 3D US prostate images.

AB - A novel statistical shape model is presented for automatic and accurate segmentation of prostate boundary from 3D ultrasound (US) images, using a hierarchical texture-based matching method. This method uses three steps. First, Gabor filter banks are used to capture rotation-invariant texture features at different scales and orientations. Second, different levels of texture features are integrated by a kernel support vector machine (KSVM) to optimally differentiate the prostate from surrounding tissues. Third, a statistical shape model is hierarchically deformed to the prostate boundary by robust texture and shape matching. Experimental results test the performance of the proposed method in segmenting 3D US prostate images.

UR - http://www.scopus.com/inward/record.url?scp=0344823746&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344823746&partnerID=8YFLogxK

M3 - Article

VL - 2878

SP - 688

EP - 696

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SN - 0302-9743

ER -