Automatic hybrid segmentation of dual contrast cardiac MR data

A. Pednekar; I. A. Kakadiaris; V. Zavaletta; R. Muthupillai; S. Flamm

doi:10.1007/3-540-45786-0_85

Automatic hybrid segmentation of dual contrast cardiac MR data

A. Pednekar, I. A. Kakadiaris, V. Zavaletta, R. Muthupillai, S. Flamm

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Scopus citations

Abstract

Manual tracing of the blood pool from short axis cine MR images is routinely used to compute ejection fraction (EF) in clinical practice. The manual segmentation process is cumbersome, time consuming, and operator dependent. In this paper, we present an algorithm for the automatic computation of the EF that is based on segmenting the left ventricle by combining the fuzzy connectedness and deformable model frameworks. Our contributions are the following: 1) we automatically estimate a seed point and sample region for the fuzzy connectedness estimates, 2) we extend the fuzzy connectedness method to use adaptive weights for the homogeneity and the gradient energy functions that are computed dynamically, and 3) we extend the hybrid segmentation framework to allow forces from dual contrast and fuzzy connectedness data integrated, with shape constraints. Finally, we compare our method against manual delineation performed by experienced radiologists on the data from nine asymptomatic volunteers with very encouraging results.

Original language	English (US)
Title of host publication	Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings
Editors	Takeyoshi Dohi, Ron Kikinis
Publisher	Springer-Verlag
Pages	690-697
Number of pages	8
ISBN (Print)	9783540457862
DOIs	https://doi.org/10.1007/3-540-45786-0_85
State	Published - 2002
Event	5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002 - Tokyo, Japan Duration: Sep 25 2002 → Sep 28 2002

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	2488
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002
Country/Territory	Japan
City	Tokyo
Period	9/25/02 → 9/28/02

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/3-540-45786-0_85

Cite this

Pednekar, A., Kakadiaris, I. A., Zavaletta, V., Muthupillai, R., & Flamm, S. (2002). Automatic hybrid segmentation of dual contrast cardiac MR data. In T. Dohi, & R. Kikinis (Eds.), Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings (pp. 690-697). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2488). Springer-Verlag. https://doi.org/10.1007/3-540-45786-0_85

Automatic hybrid segmentation of dual contrast cardiac MR data. / Pednekar, A.; Kakadiaris, I. A.; Zavaletta, V. et al.

Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. ed. / Takeyoshi Dohi; Ron Kikinis. Springer-Verlag, 2002. p. 690-697 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2488).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Pednekar, A, Kakadiaris, IA, Zavaletta, V, Muthupillai, R & Flamm, S 2002, Automatic hybrid segmentation of dual contrast cardiac MR data. in T Dohi & R Kikinis (eds), Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2488, Springer-Verlag, pp. 690-697, 5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002, Tokyo, Japan, 9/25/02. https://doi.org/10.1007/3-540-45786-0_85

Pednekar A, Kakadiaris IA, Zavaletta V, Muthupillai R, Flamm S. Automatic hybrid segmentation of dual contrast cardiac MR data. In Dohi T, Kikinis R, editors, Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. Springer-Verlag. 2002. p. 690-697. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-45786-0_85

Pednekar, A. ; Kakadiaris, I. A. ; Zavaletta, V. et al. / Automatic hybrid segmentation of dual contrast cardiac MR data. Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings. editor / Takeyoshi Dohi ; Ron Kikinis. Springer-Verlag, 2002. pp. 690-697 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{f4b64d67e2994a0fbd082063044fd892,

title = "Automatic hybrid segmentation of dual contrast cardiac MR data",

abstract = "Manual tracing of the blood pool from short axis cine MR images is routinely used to compute ejection fraction (EF) in clinical practice. The manual segmentation process is cumbersome, time consuming, and operator dependent. In this paper, we present an algorithm for the automatic computation of the EF that is based on segmenting the left ventricle by combining the fuzzy connectedness and deformable model frameworks. Our contributions are the following: 1) we automatically estimate a seed point and sample region for the fuzzy connectedness estimates, 2) we extend the fuzzy connectedness method to use adaptive weights for the homogeneity and the gradient energy functions that are computed dynamically, and 3) we extend the hybrid segmentation framework to allow forces from dual contrast and fuzzy connectedness data integrated, with shape constraints. Finally, we compare our method against manual delineation performed by experienced radiologists on the data from nine asymptomatic volunteers with very encouraging results.",

author = "A. Pednekar and Kakadiaris, {I. A.} and V. Zavaletta and R. Muthupillai and S. Flamm",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2002.; 5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002 ; Conference date: 25-09-2002 Through 28-09-2002",

year = "2002",

doi = "10.1007/3-540-45786-0_85",

language = "English (US)",

isbn = "9783540457862",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer-Verlag",

pages = "690--697",

editor = "Takeyoshi Dohi and Ron Kikinis",

booktitle = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings",

}

TY - GEN

T1 - Automatic hybrid segmentation of dual contrast cardiac MR data

AU - Pednekar, A.

AU - Kakadiaris, I. A.

AU - Zavaletta, V.

AU - Muthupillai, R.

AU - Flamm, S.

PY - 2002

Y1 - 2002

N2 - Manual tracing of the blood pool from short axis cine MR images is routinely used to compute ejection fraction (EF) in clinical practice. The manual segmentation process is cumbersome, time consuming, and operator dependent. In this paper, we present an algorithm for the automatic computation of the EF that is based on segmenting the left ventricle by combining the fuzzy connectedness and deformable model frameworks. Our contributions are the following: 1) we automatically estimate a seed point and sample region for the fuzzy connectedness estimates, 2) we extend the fuzzy connectedness method to use adaptive weights for the homogeneity and the gradient energy functions that are computed dynamically, and 3) we extend the hybrid segmentation framework to allow forces from dual contrast and fuzzy connectedness data integrated, with shape constraints. Finally, we compare our method against manual delineation performed by experienced radiologists on the data from nine asymptomatic volunteers with very encouraging results.

AB - Manual tracing of the blood pool from short axis cine MR images is routinely used to compute ejection fraction (EF) in clinical practice. The manual segmentation process is cumbersome, time consuming, and operator dependent. In this paper, we present an algorithm for the automatic computation of the EF that is based on segmenting the left ventricle by combining the fuzzy connectedness and deformable model frameworks. Our contributions are the following: 1) we automatically estimate a seed point and sample region for the fuzzy connectedness estimates, 2) we extend the fuzzy connectedness method to use adaptive weights for the homogeneity and the gradient energy functions that are computed dynamically, and 3) we extend the hybrid segmentation framework to allow forces from dual contrast and fuzzy connectedness data integrated, with shape constraints. Finally, we compare our method against manual delineation performed by experienced radiologists on the data from nine asymptomatic volunteers with very encouraging results.

UR - http://www.scopus.com/inward/record.url?scp=33646187427&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646187427&partnerID=8YFLogxK

U2 - 10.1007/3-540-45786-0_85

DO - 10.1007/3-540-45786-0_85

M3 - Conference contribution

AN - SCOPUS:33646187427

SN - 9783540457862

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 690

EP - 697

BT - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002 - 5th International Conference, Proceedings

A2 - Dohi, Takeyoshi

A2 - Kikinis, Ron

PB - Springer-Verlag

T2 - 5th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2002

Y2 - 25 September 2002 through 28 September 2002

ER -

Automatic hybrid segmentation of dual contrast cardiac MR data

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this