Multiscale Vessel-guided Airway Tree Segmentation

Pechin Chien Pau Lo; Jon Sporring; Marleen de Bruijne

Multiscale Vessel-guided Airway Tree Segmentation

Pechin Chien Pau Lo, Jon Sporring, Marleen de Bruijne

Abstract

This paper presents a method for airway tree segmentation
that uses a combination of a trained airway appearance model, vessel
and airway orientation information, and region growing. The method
uses a voxel classification based appearance model, which involves the
use of a classifier that is trained to differentiate between airway and
non-airway voxels. Vessel and airway orientation information are used in
the form of a vessel orientation similarity measure, which indicates how
similar the orientation of the an airway candidate is to the orientation
of the neighboring vessel. The method is evaluated within EXACT’09
on a diverse set of CT scans. Results show a favorable combination of a
relatively large portion of the tree detected correctly with very few false
positives.

Original language	English
Title of host publication	Proceedings of Second International Workshop on Pulmonary Image Analysis
Editors	Matthew Brown, Marleen de Bruijne, Bram van Ginneken, Atilla Kiraly, Jan-Martin Kuhnigk, Cristian Lorenz, Jamie R McClelland, Kensaku Mori, Antohony Reeves, Joseph M Reinhardt
Number of pages	10
Publisher	CreateSpace
Publication date	2009
Pages	323-332
ISBN (Print)	978-1448680894, 1448680891
Publication status	Published - 2009
Event	The Second International Workshop on Pulmonary Image Analysis - London, United Kingdom Duration: 20 Sept 2009 → 20 Sept 2009 Conference number: 2

Conference

Conference	The Second International Workshop on Pulmonary Image Analysis
Number	2
Country/Territory	United Kingdom
City	London
Period	20/09/2009 → 20/09/2009

Access to Document

http://www.lungworkshop.org/2009/proc2009/323.pdf

Cite this

Lo, P. C. P., Sporring, J., & de Bruijne, M. (2009). Multiscale Vessel-guided Airway Tree Segmentation. In M. Brown, M. de Bruijne, B. van Ginneken, A. Kiraly, J-M. Kuhnigk, C. Lorenz, J. R. McClelland, K. Mori, A. Reeves, & J. M. Reinhardt (Eds.), Proceedings of Second International Workshop on Pulmonary Image Analysis (pp. 323-332). CreateSpace. http://www.lungworkshop.org/2009/proc2009/323.pdf

Multiscale Vessel-guided Airway Tree Segmentation. / Lo, Pechin Chien Pau; Sporring, Jon ; de Bruijne, Marleen.

Proceedings of Second International Workshop on Pulmonary Image Analysis. ed. / Matthew Brown; Marleen de Bruijne; Bram van Ginneken; Atilla Kiraly; Jan-Martin Kuhnigk; Cristian Lorenz; Jamie R McClelland; Kensaku Mori; Antohony Reeves; Joseph M Reinhardt. CreateSpace, 2009. p. 323-332.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Lo, PCP, Sporring, J & de Bruijne, M 2009, Multiscale Vessel-guided Airway Tree Segmentation. in M Brown, M de Bruijne, B van Ginneken, A Kiraly, J-M Kuhnigk, C Lorenz, JR McClelland, K Mori, A Reeves & JM Reinhardt (eds), Proceedings of Second International Workshop on Pulmonary Image Analysis. CreateSpace, pp. 323-332, The Second International Workshop on Pulmonary Image Analysis, London, United Kingdom, 20/09/2009. <http://www.lungworkshop.org/2009/proc2009/323.pdf>

Lo, Pechin Chien Pau ; Sporring, Jon ; de Bruijne, Marleen. / Multiscale Vessel-guided Airway Tree Segmentation. Proceedings of Second International Workshop on Pulmonary Image Analysis. editor / Matthew Brown ; Marleen de Bruijne ; Bram van Ginneken ; Atilla Kiraly ; Jan-Martin Kuhnigk ; Cristian Lorenz ; Jamie R McClelland ; Kensaku Mori ; Antohony Reeves ; Joseph M Reinhardt. CreateSpace, 2009. pp. 323-332

@inproceedings{98738a709d3511debc73000ea68e967b,

title = "Multiscale Vessel-guided Airway Tree Segmentation",

abstract = "This paper presents a method for airway tree segmentationthat uses a combination of a trained airway appearance model, vesseland airway orientation information, and region growing. The methoduses a voxel classification based appearance model, which involves theuse of a classifier that is trained to differentiate between airway andnon-airway voxels. Vessel and airway orientation information are used inthe form of a vessel orientation similarity measure, which indicates howsimilar the orientation of the an airway candidate is to the orientationof the neighboring vessel. The method is evaluated within EXACT{\textquoteright}09on a diverse set of CT scans. Results show a favorable combination of arelatively large portion of the tree detected correctly with very few falsepositives.",

author = "Lo, {Pechin Chien Pau} and Jon Sporring and {de Bruijne}, Marleen",

year = "2009",

language = "English",

isbn = "978-1448680894",

pages = "323--332",

editor = "Matthew Brown and {de Bruijne}, Marleen and {van Ginneken}, Bram and Atilla Kiraly and Jan-Martin Kuhnigk and Cristian Lorenz and McClelland, {Jamie R} and Kensaku Mori and Antohony Reeves and Reinhardt, {Joseph M}",

booktitle = "Proceedings of Second International Workshop on Pulmonary Image Analysis",

publisher = "CreateSpace",

note = "The Second International Workshop on Pulmonary Image Analysis ; Conference date: 20-09-2009 Through 20-09-2009",

}

TY - GEN

T1 - Multiscale Vessel-guided Airway Tree Segmentation

AU - Lo, Pechin Chien Pau

AU - Sporring, Jon

AU - de Bruijne, Marleen

N1 - Conference code: 2

PY - 2009

Y1 - 2009

N2 - This paper presents a method for airway tree segmentationthat uses a combination of a trained airway appearance model, vesseland airway orientation information, and region growing. The methoduses a voxel classification based appearance model, which involves theuse of a classifier that is trained to differentiate between airway andnon-airway voxels. Vessel and airway orientation information are used inthe form of a vessel orientation similarity measure, which indicates howsimilar the orientation of the an airway candidate is to the orientationof the neighboring vessel. The method is evaluated within EXACT’09on a diverse set of CT scans. Results show a favorable combination of arelatively large portion of the tree detected correctly with very few falsepositives.

AB - This paper presents a method for airway tree segmentationthat uses a combination of a trained airway appearance model, vesseland airway orientation information, and region growing. The methoduses a voxel classification based appearance model, which involves theuse of a classifier that is trained to differentiate between airway andnon-airway voxels. Vessel and airway orientation information are used inthe form of a vessel orientation similarity measure, which indicates howsimilar the orientation of the an airway candidate is to the orientationof the neighboring vessel. The method is evaluated within EXACT’09on a diverse set of CT scans. Results show a favorable combination of arelatively large portion of the tree detected correctly with very few falsepositives.

M3 - Article in proceedings

SN - 978-1448680894

SN - 1448680891

SP - 323

EP - 332

BT - Proceedings of Second International Workshop on Pulmonary Image Analysis

A2 - Brown, Matthew

A2 - de Bruijne, Marleen

A2 - van Ginneken, Bram

A2 - Kiraly, Atilla

A2 - Kuhnigk, Jan-Martin

A2 - Lorenz, Cristian

A2 - McClelland, Jamie R

A2 - Mori, Kensaku

A2 - Reeves, Antohony

A2 - Reinhardt, Joseph M

PB - CreateSpace

T2 - The Second International Workshop on Pulmonary Image Analysis

Y2 - 20 September 2009 through 20 September 2009

ER -

Multiscale Vessel-guided Airway Tree Segmentation

Abstract

Conference

Access to Document

Fingerprint

Cite this