Supervised method to build an atlas database for multi-atlas segmentation-propagation

Kaikai Shen; Pierrick Bourgeat; Jurgen Fripp; Fabrice Meriaudeau; David Ames; Kathryn A. Ellis; Colin L. Masters; Victor L. Villemagne; Christopher C. Rowe; Olivier Salvado

doi:10.1117/12.844048

Supervised method to build an atlas database for multi-atlas segmentation-propagation

Kaikai Shen^*, Pierrick Bourgeat, Jurgen Fripp, Fabrice Meriaudeau, David Ames, Kathryn A. Ellis, Colin L. Masters, Victor L. Villemagne, Christopher C. Rowe, Olivier Salvado

^*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Multi-atlas based segmentation-propagation approaches have been shown to obtain accurate parcelation of brain structures. However, this approach requires a large number of manually delineated atlases, which are often not available. We propose a supervised method to build a population specific atlas database, using the publicly available Internet Brain Segmentation Repository (IBSR). The set of atlases grows iteratively as new atlases are added, so that its segmentation capability may be enhanced in the multi-atlas based approach. Using a dataset of 210 MR images of elderly subjects (170 elderly controls, 40 Alzheimer's disease) from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study, 40 MR images were segmented to build a population specific atlas database for the purpose of multi-atlas segmentation-propagation. The population specific atlases were used to segment the elderly population of 210 MR images, and were evaluated in terms of the agreement among the propagated labels. The agreement was measured by using the entropy H of the probability image produced when fused by voting rule and the partial moment μ₂ of the histogram. Compared with using IBSR atlases, the population specific atlases obtained a higher agreement when dealing with images of elderly subjects.

Original language	English
Title of host publication	Medical Imaging 2010
Subtitle of host publication	Computer-Aided Diagnosis
Editors	Nico Karssemeijer, Ronald M. Summers
Place of Publication	Washington
Publisher	SPIE
Pages	1-8
Number of pages	8
ISBN (Electronic)	9780819480255
DOIs	https://doi.org/10.1117/12.844048
Publication status	Published - 2010
Externally published	Yes
Event	Conference on Medical Imaging 2010 - Computer-Aided Diagnosis - San Diego, Canada Duration: 16 Feb 2010 → 18 Feb 2010

Publication series

Name	Proceedings of SPIE
Publisher	SPIE-INT SOC OPTICAL ENGINEERING
Volume	7624
ISSN (Print)	0277-786X

Conference

Conference	Conference on Medical Imaging 2010 - Computer-Aided Diagnosis
Country/Territory	Canada
City	San Diego
Period	16/02/10 → 18/02/10

Keywords

Image segmentation
multi-atlas segmentation-propagation
MRI

Access to Document

10.1117/12.844048

Cite this

Shen, K., Bourgeat, P., Fripp, J., Meriaudeau, F., Ames, D., Ellis, K. A., Masters, C. L., Villemagne, V. L., Rowe, C. C., & Salvado, O. (2010). Supervised method to build an atlas database for multi-atlas segmentation-propagation. In N. Karssemeijer, & R. M. Summers (Eds.), Medical Imaging 2010: Computer-Aided Diagnosis (pp. 1-8). [76241N] (Proceedings of SPIE; Vol. 7624). SPIE. https://doi.org/10.1117/12.844048

@inproceedings{28a13cadbf8d4e91a0532f7ec336394c,

title = "Supervised method to build an atlas database for multi-atlas segmentation-propagation",

abstract = "Multi-atlas based segmentation-propagation approaches have been shown to obtain accurate parcelation of brain structures. However, this approach requires a large number of manually delineated atlases, which are often not available. We propose a supervised method to build a population specific atlas database, using the publicly available Internet Brain Segmentation Repository (IBSR). The set of atlases grows iteratively as new atlases are added, so that its segmentation capability may be enhanced in the multi-atlas based approach. Using a dataset of 210 MR images of elderly subjects (170 elderly controls, 40 Alzheimer's disease) from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study, 40 MR images were segmented to build a population specific atlas database for the purpose of multi-atlas segmentation-propagation. The population specific atlases were used to segment the elderly population of 210 MR images, and were evaluated in terms of the agreement among the propagated labels. The agreement was measured by using the entropy H of the probability image produced when fused by voting rule and the partial moment μ2 of the histogram. Compared with using IBSR atlases, the population specific atlases obtained a higher agreement when dealing with images of elderly subjects.",

keywords = "Image segmentation, multi-atlas segmentation-propagation, MRI",

author = "Kaikai Shen and Pierrick Bourgeat and Jurgen Fripp and Fabrice Meriaudeau and David Ames and Ellis, {Kathryn A.} and Masters, {Colin L.} and Villemagne, {Victor L.} and Rowe, {Christopher C.} and Olivier Salvado",

year = "2010",

doi = "10.1117/12.844048",

language = "English",

series = "Proceedings of SPIE",

publisher = "SPIE",

pages = "1--8",

editor = "Nico Karssemeijer and Summers, {Ronald M.}",

booktitle = "Medical Imaging 2010",

address = "United States",

note = "Conference on Medical Imaging 2010 - Computer-Aided Diagnosis ; Conference date: 16-02-2010 Through 18-02-2010",

}

Shen, K, Bourgeat, P, Fripp, J, Meriaudeau, F, Ames, D, Ellis, KA, Masters, CL, Villemagne, VL, Rowe, CC & Salvado, O 2010, Supervised method to build an atlas database for multi-atlas segmentation-propagation. in N Karssemeijer & RM Summers (eds), Medical Imaging 2010: Computer-Aided Diagnosis., 76241N, Proceedings of SPIE, vol. 7624, SPIE, Washington, pp. 1-8, Conference on Medical Imaging 2010 - Computer-Aided Diagnosis, San Diego, Canada, 16/02/10. https://doi.org/10.1117/12.844048

Supervised method to build an atlas database for multi-atlas segmentation-propagation. / Shen, Kaikai; Bourgeat, Pierrick; Fripp, Jurgen et al.
Medical Imaging 2010: Computer-Aided Diagnosis. ed. / Nico Karssemeijer; Ronald M. Summers. Washington: SPIE, 2010. p. 1-8 76241N (Proceedings of SPIE; Vol. 7624).

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

TY - GEN

T1 - Supervised method to build an atlas database for multi-atlas segmentation-propagation

AU - Shen, Kaikai

AU - Bourgeat, Pierrick

AU - Fripp, Jurgen

AU - Meriaudeau, Fabrice

AU - Ames, David

AU - Ellis, Kathryn A.

AU - Masters, Colin L.

AU - Villemagne, Victor L.

AU - Rowe, Christopher C.

AU - Salvado, Olivier

PY - 2010

Y1 - 2010

N2 - Multi-atlas based segmentation-propagation approaches have been shown to obtain accurate parcelation of brain structures. However, this approach requires a large number of manually delineated atlases, which are often not available. We propose a supervised method to build a population specific atlas database, using the publicly available Internet Brain Segmentation Repository (IBSR). The set of atlases grows iteratively as new atlases are added, so that its segmentation capability may be enhanced in the multi-atlas based approach. Using a dataset of 210 MR images of elderly subjects (170 elderly controls, 40 Alzheimer's disease) from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study, 40 MR images were segmented to build a population specific atlas database for the purpose of multi-atlas segmentation-propagation. The population specific atlases were used to segment the elderly population of 210 MR images, and were evaluated in terms of the agreement among the propagated labels. The agreement was measured by using the entropy H of the probability image produced when fused by voting rule and the partial moment μ2 of the histogram. Compared with using IBSR atlases, the population specific atlases obtained a higher agreement when dealing with images of elderly subjects.

AB - Multi-atlas based segmentation-propagation approaches have been shown to obtain accurate parcelation of brain structures. However, this approach requires a large number of manually delineated atlases, which are often not available. We propose a supervised method to build a population specific atlas database, using the publicly available Internet Brain Segmentation Repository (IBSR). The set of atlases grows iteratively as new atlases are added, so that its segmentation capability may be enhanced in the multi-atlas based approach. Using a dataset of 210 MR images of elderly subjects (170 elderly controls, 40 Alzheimer's disease) from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study, 40 MR images were segmented to build a population specific atlas database for the purpose of multi-atlas segmentation-propagation. The population specific atlases were used to segment the elderly population of 210 MR images, and were evaluated in terms of the agreement among the propagated labels. The agreement was measured by using the entropy H of the probability image produced when fused by voting rule and the partial moment μ2 of the histogram. Compared with using IBSR atlases, the population specific atlases obtained a higher agreement when dealing with images of elderly subjects.

KW - Image segmentation

KW - multi-atlas segmentation-propagation

KW - MRI

U2 - 10.1117/12.844048

DO - 10.1117/12.844048

M3 - Conference proceeding contribution

T3 - Proceedings of SPIE

SP - 1

EP - 8

BT - Medical Imaging 2010

A2 - Karssemeijer, Nico

A2 - Summers, Ronald M.

PB - SPIE

CY - Washington

T2 - Conference on Medical Imaging 2010 - Computer-Aided Diagnosis

Y2 - 16 February 2010 through 18 February 2010

ER -

Supervised method to build an atlas database for multi-atlas segmentation-propagation

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this