A texture feature-based method for dynamic organ tracking

Zhen Tian, Caijie Duan, Kehong Yuan, Wei Han, Datian Ye

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this paper, a novel approach for dynamic organ tracking based on texture feature is proposed. The purpose of this work is to find an efficient way to delineate and track the contours of a specific organ such as mouth, eye and lung which has distinct iiiteiiRity from the nrroniidiiigR aiid whoRe Rhape arid poRitiori vary Rmoothiy. The initial organ contours were manually delineated as the reference. Control markers on the organ contours were chosen for automatically tracking the dynamic organ contours. The tracking procedure consists of two major steps. Firstly, texture features were extracted from the regions under these control markers by using Sobel operator; and secondly, a local searching strategy based on the texture feature was performed for marker matching and further the dynamic contours tracking. The advantage of this approach lies on that it makes use of the local contrast between the organ and its surroundings to capture efficient local texture features and match fast to the corresponding location on target images while considering the motion and deformation of the organ in target images. The proposed approach was tested by tracking eye, mouth contours in videos and lung contours in clinical 4D thoracic CT images respectively. The satisfied results were obtained. An accordance coefficient was proposed to quantitatively evaluate the tracking performance and it was found that our approach performed best in lung contour tracking with accordance coefficients about 95%.

Original languageEnglish (US)
Pages (from-to)5697-5708
Number of pages12
JournalInternational Journal of Innovative Computing, Information and Control
Volume6
Issue number12
StatePublished - Dec 2010

Fingerprint

Texture Feature
Textures
Lung
Eye Tracking
Target
CT Image
Coefficient
Vary
Distinct
Motion
Evaluate
Operator

Keywords

  • Accordance coefficient
  • Dynamic contour tracking
  • Sobel operator
  • Texture feature

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this

A texture feature-based method for dynamic organ tracking. / Tian, Zhen; Duan, Caijie; Yuan, Kehong; Han, Wei; Ye, Datian.

In: International Journal of Innovative Computing, Information and Control, Vol. 6, No. 12, 12.2010, p. 5697-5708.

Research output: Contribution to journalArticle

Tian, Zhen ; Duan, Caijie ; Yuan, Kehong ; Han, Wei ; Ye, Datian. / A texture feature-based method for dynamic organ tracking. In: International Journal of Innovative Computing, Information and Control. 2010 ; Vol. 6, No. 12. pp. 5697-5708.
@article{4ca93344d8c141a7ae0a3379e0d195e9,
title = "A texture feature-based method for dynamic organ tracking",
abstract = "In this paper, a novel approach for dynamic organ tracking based on texture feature is proposed. The purpose of this work is to find an efficient way to delineate and track the contours of a specific organ such as mouth, eye and lung which has distinct iiiteiiRity from the nrroniidiiigR aiid whoRe Rhape arid poRitiori vary Rmoothiy. The initial organ contours were manually delineated as the reference. Control markers on the organ contours were chosen for automatically tracking the dynamic organ contours. The tracking procedure consists of two major steps. Firstly, texture features were extracted from the regions under these control markers by using Sobel operator; and secondly, a local searching strategy based on the texture feature was performed for marker matching and further the dynamic contours tracking. The advantage of this approach lies on that it makes use of the local contrast between the organ and its surroundings to capture efficient local texture features and match fast to the corresponding location on target images while considering the motion and deformation of the organ in target images. The proposed approach was tested by tracking eye, mouth contours in videos and lung contours in clinical 4D thoracic CT images respectively. The satisfied results were obtained. An accordance coefficient was proposed to quantitatively evaluate the tracking performance and it was found that our approach performed best in lung contour tracking with accordance coefficients about 95{\%}.",
keywords = "Accordance coefficient, Dynamic contour tracking, Sobel operator, Texture feature",
author = "Zhen Tian and Caijie Duan and Kehong Yuan and Wei Han and Datian Ye",
year = "2010",
month = "12",
language = "English (US)",
volume = "6",
pages = "5697--5708",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "12",

}

TY - JOUR

T1 - A texture feature-based method for dynamic organ tracking

AU - Tian, Zhen

AU - Duan, Caijie

AU - Yuan, Kehong

AU - Han, Wei

AU - Ye, Datian

PY - 2010/12

Y1 - 2010/12

N2 - In this paper, a novel approach for dynamic organ tracking based on texture feature is proposed. The purpose of this work is to find an efficient way to delineate and track the contours of a specific organ such as mouth, eye and lung which has distinct iiiteiiRity from the nrroniidiiigR aiid whoRe Rhape arid poRitiori vary Rmoothiy. The initial organ contours were manually delineated as the reference. Control markers on the organ contours were chosen for automatically tracking the dynamic organ contours. The tracking procedure consists of two major steps. Firstly, texture features were extracted from the regions under these control markers by using Sobel operator; and secondly, a local searching strategy based on the texture feature was performed for marker matching and further the dynamic contours tracking. The advantage of this approach lies on that it makes use of the local contrast between the organ and its surroundings to capture efficient local texture features and match fast to the corresponding location on target images while considering the motion and deformation of the organ in target images. The proposed approach was tested by tracking eye, mouth contours in videos and lung contours in clinical 4D thoracic CT images respectively. The satisfied results were obtained. An accordance coefficient was proposed to quantitatively evaluate the tracking performance and it was found that our approach performed best in lung contour tracking with accordance coefficients about 95%.

AB - In this paper, a novel approach for dynamic organ tracking based on texture feature is proposed. The purpose of this work is to find an efficient way to delineate and track the contours of a specific organ such as mouth, eye and lung which has distinct iiiteiiRity from the nrroniidiiigR aiid whoRe Rhape arid poRitiori vary Rmoothiy. The initial organ contours were manually delineated as the reference. Control markers on the organ contours were chosen for automatically tracking the dynamic organ contours. The tracking procedure consists of two major steps. Firstly, texture features were extracted from the regions under these control markers by using Sobel operator; and secondly, a local searching strategy based on the texture feature was performed for marker matching and further the dynamic contours tracking. The advantage of this approach lies on that it makes use of the local contrast between the organ and its surroundings to capture efficient local texture features and match fast to the corresponding location on target images while considering the motion and deformation of the organ in target images. The proposed approach was tested by tracking eye, mouth contours in videos and lung contours in clinical 4D thoracic CT images respectively. The satisfied results were obtained. An accordance coefficient was proposed to quantitatively evaluate the tracking performance and it was found that our approach performed best in lung contour tracking with accordance coefficients about 95%.

KW - Accordance coefficient

KW - Dynamic contour tracking

KW - Sobel operator

KW - Texture feature

UR - http://www.scopus.com/inward/record.url?scp=78650277964&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650277964&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:78650277964

VL - 6

SP - 5697

EP - 5708

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 12

ER -