Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior

Wu Qiu; Jing Yuan; Jessica Kishimoto; Yimin Chen; Martin Rajchl; Eranga Ukwatta; Sandrine de Ribaupierre; Aaron Fenster

doi:10.1007/978-3-319-24574-4_11

Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior

Wu Qiu, Jing Yuan, Jessica Kishimoto, Yimin Chen, Martin Rajchl, Eranga Ukwatta, Sandrine de Ribaupierre, Aaron Fenster

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

5 Citations (Scopus)

Abstract

Pre-term neonates born with a low birth weight (<1500g) are at increased risk for developing intraventricular hemorrhage (IVH). 3D ultrasound (US) imaging has been used to quantitatively monitor the ventricular volume in IVH neonates, instead of typical 2D US used clinically, which relies on linear measurements from a single slice and visually estimates to determine ventricular dilation. To translate 3D US imaging into clinical setting, an accurate segmentation algorithm would be desirable to automatically extract the ventricular system from 3D US images. In this paper, we propose an automatic multi-region segmentation approach for delineating lateral ventricles of pre-term neonates from 3D US images, which makes use of multi-phase geodesic level-sets (MP-GLS) segmentation technique via a variational region competition principle and a spatial shape prior derived from pre-segmented atlases. Experimental results using 15 IVH patient images show that the proposed GPU-implemented approach is accurate in terms of the Dice similarity coefficient (DSC), the mean absolute surface distance (MAD), and maximum absolute surface distance (MAXD). To the best of our knowledge, this paper reports the first study on automatic segmentation of ventricular system of premature neonatal brains from 3D US images.

Original language	English
Title of host publication	Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015
Subtitle of host publication	18th International Conference 2015 Proceedings, Part III
Editors	Nassir Navab, Joachim Hornegger, William M. Wells, Alejandro F. Frangi
Publisher	Springer
Pages	89-96
ISBN (Electronic)	9783319245744
ISBN (Print)	9783319245737
DOIs	https://doi.org/10.1007/978-3-319-24574-4_11
Publication status	Published - Oct 2015
Event	18th International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2015) - Munich, Germany Duration: 5 Oct 2015 → 9 Oct 2015 https://www.miccai2015.org/frontend/index.php?sub=22

Publication series

Name	Lecture Notes in Computer Science
Publisher	SPRINGER INT PUBLISHING AG
Volume	9351
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	18th International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2015)
Abbreviated title	MICCAI 2015
Place	Germany
City	Munich
Period	5/10/15 → 9/10/15
Internet address	https://www.miccai2015.org/frontend/index.php?sub=22

Research Keywords

3D ultrasound
pre-term neonatal ventricle segmentation
multi-phase geodesic level-sets
shape prior
convex optimization

Access Output

10.1007/978-3-319-24574-4_11

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Qiu, W., Yuan, J., Kishimoto, J., Chen, Y., Rajchl, M., Ukwatta, E., de Ribaupierre, S., & Fenster, A. (2015). Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior. In N. Navab, J. Hornegger, W. M. Wells, & A. F. Frangi (Eds.), Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference 2015 Proceedings, Part III (pp. 89-96). (Lecture Notes in Computer Science; Vol. 9351). Springer . https://doi.org/10.1007/978-3-319-24574-4_11

Qiu, Wu ; Yuan, Jing ; Kishimoto, Jessica et al. / Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference 2015 Proceedings, Part III. editor / Nassir Navab ; Joachim Hornegger ; William M. Wells ; Alejandro F. Frangi. Springer , 2015. pp. 89-96 (Lecture Notes in Computer Science).

@inproceedings{1be787cf8d874ced99dfc492b0dd9982,

title = "Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior",

abstract = "Pre-term neonates born with a low birth weight (<1500g) are at increased risk for developing intraventricular hemorrhage (IVH). 3D ultrasound (US) imaging has been used to quantitatively monitor the ventricular volume in IVH neonates, instead of typical 2D US used clinically, which relies on linear measurements from a single slice and visually estimates to determine ventricular dilation. To translate 3D US imaging into clinical setting, an accurate segmentation algorithm would be desirable to automatically extract the ventricular system from 3D US images. In this paper, we propose an automatic multi-region segmentation approach for delineating lateral ventricles of pre-term neonates from 3D US images, which makes use of multi-phase geodesic level-sets (MP-GLS) segmentation technique via a variational region competition principle and a spatial shape prior derived from pre-segmented atlases. Experimental results using 15 IVH patient images show that the proposed GPU-implemented approach is accurate in terms of the Dice similarity coefficient (DSC), the mean absolute surface distance (MAD), and maximum absolute surface distance (MAXD). To the best of our knowledge, this paper reports the first study on automatic segmentation of ventricular system of premature neonatal brains from 3D US images.",

keywords = "3D ultrasound, pre-term neonatal ventricle segmentation, multi-phase geodesic level-sets, shape prior, convex optimization",

author = "Wu Qiu and Jing Yuan and Jessica Kishimoto and Yimin Chen and Martin Rajchl and Eranga Ukwatta and {de Ribaupierre}, Sandrine and Aaron Fenster",

year = "2015",

month = oct,

doi = "10.1007/978-3-319-24574-4_11",

language = "English",

isbn = "9783319245737",

series = "Lecture Notes in Computer Science",

publisher = "Springer ",

pages = "89--96",

editor = "Nassir Navab and Joachim Hornegger and Wells, {William M.} and Frangi, {Alejandro F.}",

booktitle = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015",

note = "18th International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2015), MICCAI 2015 ; Conference date: 05-10-2015 Through 09-10-2015",

url = "https://www.miccai2015.org/frontend/index.php?sub=22",

}

Qiu, W, Yuan, J, Kishimoto, J, Chen, Y, Rajchl, M, Ukwatta, E, de Ribaupierre, S & Fenster, A 2015, Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior. in N Navab, J Hornegger, WM Wells & AF Frangi (eds), Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference 2015 Proceedings, Part III. Lecture Notes in Computer Science, vol. 9351, Springer , pp. 89-96, 18th International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2015), Munich, Germany, 5/10/15. https://doi.org/10.1007/978-3-319-24574-4_11

Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior. / Qiu, Wu; Yuan, Jing; Kishimoto, Jessica et al.
Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference 2015 Proceedings, Part III. ed. / Nassir Navab; Joachim Hornegger; William M. Wells; Alejandro F. Frangi. Springer , 2015. p. 89-96 (Lecture Notes in Computer Science; Vol. 9351).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior

AU - Qiu, Wu

AU - Yuan, Jing

AU - Kishimoto, Jessica

AU - Chen, Yimin

AU - Rajchl, Martin

AU - Ukwatta, Eranga

AU - de Ribaupierre, Sandrine

AU - Fenster, Aaron

PY - 2015/10

Y1 - 2015/10

N2 - Pre-term neonates born with a low birth weight (<1500g) are at increased risk for developing intraventricular hemorrhage (IVH). 3D ultrasound (US) imaging has been used to quantitatively monitor the ventricular volume in IVH neonates, instead of typical 2D US used clinically, which relies on linear measurements from a single slice and visually estimates to determine ventricular dilation. To translate 3D US imaging into clinical setting, an accurate segmentation algorithm would be desirable to automatically extract the ventricular system from 3D US images. In this paper, we propose an automatic multi-region segmentation approach for delineating lateral ventricles of pre-term neonates from 3D US images, which makes use of multi-phase geodesic level-sets (MP-GLS) segmentation technique via a variational region competition principle and a spatial shape prior derived from pre-segmented atlases. Experimental results using 15 IVH patient images show that the proposed GPU-implemented approach is accurate in terms of the Dice similarity coefficient (DSC), the mean absolute surface distance (MAD), and maximum absolute surface distance (MAXD). To the best of our knowledge, this paper reports the first study on automatic segmentation of ventricular system of premature neonatal brains from 3D US images.

AB - Pre-term neonates born with a low birth weight (<1500g) are at increased risk for developing intraventricular hemorrhage (IVH). 3D ultrasound (US) imaging has been used to quantitatively monitor the ventricular volume in IVH neonates, instead of typical 2D US used clinically, which relies on linear measurements from a single slice and visually estimates to determine ventricular dilation. To translate 3D US imaging into clinical setting, an accurate segmentation algorithm would be desirable to automatically extract the ventricular system from 3D US images. In this paper, we propose an automatic multi-region segmentation approach for delineating lateral ventricles of pre-term neonates from 3D US images, which makes use of multi-phase geodesic level-sets (MP-GLS) segmentation technique via a variational region competition principle and a spatial shape prior derived from pre-segmented atlases. Experimental results using 15 IVH patient images show that the proposed GPU-implemented approach is accurate in terms of the Dice similarity coefficient (DSC), the mean absolute surface distance (MAD), and maximum absolute surface distance (MAXD). To the best of our knowledge, this paper reports the first study on automatic segmentation of ventricular system of premature neonatal brains from 3D US images.

KW - 3D ultrasound

KW - pre-term neonatal ventricle segmentation

KW - multi-phase geodesic level-sets

KW - shape prior

KW - convex optimization

UR - http://www.scopus.com/inward/record.url?scp=84951776427&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84951776427&origin=recordpage

U2 - 10.1007/978-3-319-24574-4_11

DO - 10.1007/978-3-319-24574-4_11

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9783319245737

T3 - Lecture Notes in Computer Science

SP - 89

EP - 96

BT - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015

A2 - Navab, Nassir

A2 - Hornegger, Joachim

A2 - Wells, William M.

A2 - Frangi, Alejandro F.

PB - Springer

T2 - 18th International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2015)

Y2 - 5 October 2015 through 9 October 2015

ER -

Qiu W, Yuan J, Kishimoto J, Chen Y, Rajchl M, Ukwatta E et al. Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior. In Navab N, Hornegger J, Wells WM, Frangi AF, editors, Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015: 18th International Conference 2015 Proceedings, Part III. Springer . 2015. p. 89-96. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-24574-4_11

Automatic 3D US Brain Ventricle Segmentation in Pre-Term Neonates Using Multi-phase Geodesic Level-Sets with Shape Prior

Abstract

Publication series

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this