Segmenting the Parotid Gland using Registration and Level Set Methods

Christian Hollensen; Mads Fogtmann Hansen; Liselotte Højgaard; Lena Specht; Rasmus Larsen

Segmenting the Parotid Gland using Registration and Level Set Methods

Christian Hollensen, Mads Fogtmann Hansen, Liselotte Højgaard, Lena Specht, Rasmus Larsen

Copenhagen University Hospital Herlev and Gentofte

Research output: Contribution to conference › Poster › Research › peer-review

Abstract

The bilateral parotid glands were segmented using a registration scheme followed by level set segmentation. A training set consisting of computerized tomography from 10 patients with segmentation of the bilateral glands were used to optimize the parameters of registration and level set segmentation. The method was evaluated on a test set consisting of 8 corresponding data sets. The attained total volume Dice coefficient and mean Haussdorff distance were 0.61 ± 0.20 and 15.6 ± 7.4 mm respectively. The method has improvement potential which could be exploited in order for clinical introduction.

Original language	English
Publication date	2010
Publication status	Published - 2010
Event	13th International Conference on Medical Image Computing and Computer Assisted Intervention - Beijing, China Duration: 20 Sept 2010 → 24 Sept 2010 Conference number: 13 http://www.miccai2010.org/

Conference

Conference	13th International Conference on Medical Image Computing and Computer Assisted Intervention
Number	13
Country/Territory	China
City	Beijing
Period	20/09/2010 → 24/09/2010
Internet address	http://www.miccai2010.org/

Keywords

Registration
CT
Segmentation
Head&Neck

OpenUrl availability

Full text

Cite this

@conference{a188213f48f14870b243275b60bd6965,

title = "Segmenting the Parotid Gland using Registration and Level Set Methods",

abstract = "The bilateral parotid glands were segmented using a registration scheme followed by level set segmentation. A training set consisting of computerized tomography from 10 patients with segmentation of the bilateral glands were used to optimize the parameters of registration and level set segmentation. The method was evaluated on a test set consisting of 8 corresponding data sets. The attained total volume Dice coefficient and mean Haussdorff distance were 0.61 ± 0.20 and 15.6 ± 7.4 mm respectively. The method has improvement potential which could be exploited in order for clinical introduction.",

keywords = "Registration, CT, Segmentation, Head&Neck",

author = "Christian Hollensen and Hansen, {Mads Fogtmann} and Liselotte H{\o}jgaard and Lena Specht and Rasmus Larsen",

year = "2010",

language = "English",

note = "13th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI2010 ; Conference date: 20-09-2010 Through 24-09-2010",

url = "http://www.miccai2010.org/",

}

Segmenting the Parotid Gland using Registration and Level Set Methods. / Hollensen, Christian; Hansen, Mads Fogtmann; Højgaard, Liselotte et al.
2010. Poster session presented at 13th International Conference on Medical Image Computing and Computer Assisted Intervention, Beijing, China.

Research output: Contribution to conference › Poster › Research › peer-review

TY - CONF

T1 - Segmenting the Parotid Gland using Registration and Level Set Methods

AU - Hollensen, Christian

AU - Hansen, Mads Fogtmann

AU - Højgaard, Liselotte

AU - Specht, Lena

AU - Larsen, Rasmus

N1 - Conference code: 13

PY - 2010

Y1 - 2010

N2 - The bilateral parotid glands were segmented using a registration scheme followed by level set segmentation. A training set consisting of computerized tomography from 10 patients with segmentation of the bilateral glands were used to optimize the parameters of registration and level set segmentation. The method was evaluated on a test set consisting of 8 corresponding data sets. The attained total volume Dice coefficient and mean Haussdorff distance were 0.61 ± 0.20 and 15.6 ± 7.4 mm respectively. The method has improvement potential which could be exploited in order for clinical introduction.

AB - The bilateral parotid glands were segmented using a registration scheme followed by level set segmentation. A training set consisting of computerized tomography from 10 patients with segmentation of the bilateral glands were used to optimize the parameters of registration and level set segmentation. The method was evaluated on a test set consisting of 8 corresponding data sets. The attained total volume Dice coefficient and mean Haussdorff distance were 0.61 ± 0.20 and 15.6 ± 7.4 mm respectively. The method has improvement potential which could be exploited in order for clinical introduction.

KW - Registration

KW - CT

KW - Segmentation

KW - Head&Neck

M3 - Poster

T2 - 13th International Conference on Medical Image Computing and Computer Assisted Intervention

Y2 - 20 September 2010 through 24 September 2010

ER -

Segmenting the Parotid Gland using Registration and Level Set Methods

Abstract

Conference

Keywords

OpenUrl availability

Fingerprint

Cite this