Coronary Lumen Segmentation Using Graph Cuts and Robust Kernel Regression

Michiel Schaap; Lisan Neefjes; Coert Metz; Alina Giessen; A.C. Weustink; Nico Mollet; Jolanda Wentzel; Theo van Walsum; Wiro Niessen

doi:10.1007/978-3-642-02498-6_44

Coronary Lumen Segmentation Using Graph Cuts and Robust Kernel Regression

Michiel Schaap, Lisan Neefjes, Coert Metz, Alina Giessen, A.C. Weustink, Nico Mollet, Jolanda Wentzel, Theo van Walsum, Wiro Niessen

Research output: Contribution to journal › Article › Academic

51 Citations (Scopus)

Abstract

This paper presents a novel method for segmenting the coronary lumen in CTA data. The method is based on graph cuts, with edge-weights depending on the intensity of the centerline, and robust kernel regression. A quantitative evaluation in 28 coronary arteries from 12 patients is performed by comparing the semi-automatic segmentations to manual annotations. This evaluation showed that the method was able to segment the coronary arteries with high accuracy, compared to manually annotated segmentations, which is reflected in a Dice coefficient of 0.85 and average symmetric surface distance of 0.22 mm.

Original language	Undefined/Unknown
Pages (from-to)	528-539
Number of pages	12
Journal	Lecture Notes in Computer Science
Volume	5636
DOIs	https://doi.org/10.1007/978-3-642-02498-6_44
Publication status	Published - 2009

Access to Document

10.1007/978-3-642-02498-6_44

http://hdl.handle.net/1765/17397

Cite this

@article{f2ea4ab1f3c946f0b26d7390166e0961,

title = "Coronary Lumen Segmentation Using Graph Cuts and Robust Kernel Regression",

abstract = "This paper presents a novel method for segmenting the coronary lumen in CTA data. The method is based on graph cuts, with edge-weights depending on the intensity of the centerline, and robust kernel regression. A quantitative evaluation in 28 coronary arteries from 12 patients is performed by comparing the semi-automatic segmentations to manual annotations. This evaluation showed that the method was able to segment the coronary arteries with high accuracy, compared to manually annotated segmentations, which is reflected in a Dice coefficient of 0.85 and average symmetric surface distance of 0.22 mm.",

author = "Michiel Schaap and Lisan Neefjes and Coert Metz and Alina Giessen and A.C. Weustink and Nico Mollet and Jolanda Wentzel and {van Walsum}, Theo and Wiro Niessen",

year = "2009",

doi = "10.1007/978-3-642-02498-6_44",

language = "Undefined/Unknown",

volume = "5636",

pages = "528--539",

journal = "Lecture Notes in Computer Science",

issn = "0302-9743",

publisher = "Springer-Verlag",

}

TY - JOUR

T1 - Coronary Lumen Segmentation Using Graph Cuts and Robust Kernel Regression

AU - Schaap, Michiel

AU - Neefjes, Lisan

AU - Metz, Coert

AU - Giessen, Alina

AU - Weustink, A.C.

AU - Mollet, Nico

AU - Wentzel, Jolanda

AU - van Walsum, Theo

AU - Niessen, Wiro

PY - 2009

Y1 - 2009

N2 - This paper presents a novel method for segmenting the coronary lumen in CTA data. The method is based on graph cuts, with edge-weights depending on the intensity of the centerline, and robust kernel regression. A quantitative evaluation in 28 coronary arteries from 12 patients is performed by comparing the semi-automatic segmentations to manual annotations. This evaluation showed that the method was able to segment the coronary arteries with high accuracy, compared to manually annotated segmentations, which is reflected in a Dice coefficient of 0.85 and average symmetric surface distance of 0.22 mm.

AB - This paper presents a novel method for segmenting the coronary lumen in CTA data. The method is based on graph cuts, with edge-weights depending on the intensity of the centerline, and robust kernel regression. A quantitative evaluation in 28 coronary arteries from 12 patients is performed by comparing the semi-automatic segmentations to manual annotations. This evaluation showed that the method was able to segment the coronary arteries with high accuracy, compared to manually annotated segmentations, which is reflected in a Dice coefficient of 0.85 and average symmetric surface distance of 0.22 mm.

U2 - 10.1007/978-3-642-02498-6_44

DO - 10.1007/978-3-642-02498-6_44

M3 - Article

SN - 0302-9743

VL - 5636

SP - 528

EP - 539

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

ER -