Neonatal brain MRI segmentation by building multi-region-multi-reference atlases

Feng Shi, Pew Thian Yap, Yong Fan, John H. Gilmore, Weili Lin, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Neonatal brain MRI segmentation is challenging due to the poor image quality. Existing population atlases used for guiding segmentation are usually constructed by averaging all images in a population with no preference. However, such approaches diminish the important local inter-subject structural variability. In this paper, we propose a multiregion- multi-reference strategy for atlas building from a population. In brief, the brain is first parcellated into multiple anatomical regions, and for each region, the population images are classified into different subpopulations. The exemplars in sub-populations serve as structural references when determining the most suitable regional atlas for a to-be-segmented image. A final atlas is generated by combining all selected regional atlases, and a joint registration-segmentation strategy is employed for tissue segmentation. Experimental results demonstrate that segmentation with our atlas achieves high average tissue overlap rates with manual golden standard of 0.86 (SD 0.02) for gray matter (GM) and 0.83 (SD 0.03) for white matter (WM), and outperforms other atlases in comparison.

Original languageEnglish
Title of host publication2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings
Pages964-967
Number of pages4
DOIs
Publication statusPublished - 2010 Aug 9
Externally publishedYes
Event7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Rotterdam, Netherlands
Duration: 2010 Apr 142010 Apr 17

Other

Other7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010
CountryNetherlands
CityRotterdam
Period10/4/1410/4/17

Fingerprint

Atlases
Magnetic resonance imaging
Brain
Tissue
Image quality
Population
Joints

Keywords

  • Joint registration-segmentation
  • Multiple atlases
  • Neonatal imaging
  • Tissue segmentation

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Shi, F., Yap, P. T., Fan, Y., Gilmore, J. H., Lin, W., & Shen, D. (2010). Neonatal brain MRI segmentation by building multi-region-multi-reference atlases. In 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings (pp. 964-967). [5490148] https://doi.org/10.1109/ISBI.2010.5490148

Neonatal brain MRI segmentation by building multi-region-multi-reference atlases. / Shi, Feng; Yap, Pew Thian; Fan, Yong; Gilmore, John H.; Lin, Weili; Shen, Dinggang.

2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings. 2010. p. 964-967 5490148.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shi, F, Yap, PT, Fan, Y, Gilmore, JH, Lin, W & Shen, D 2010, Neonatal brain MRI segmentation by building multi-region-multi-reference atlases. in 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings., 5490148, pp. 964-967, 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010, Rotterdam, Netherlands, 10/4/14. https://doi.org/10.1109/ISBI.2010.5490148
Shi F, Yap PT, Fan Y, Gilmore JH, Lin W, Shen D. Neonatal brain MRI segmentation by building multi-region-multi-reference atlases. In 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings. 2010. p. 964-967. 5490148 https://doi.org/10.1109/ISBI.2010.5490148
Shi, Feng ; Yap, Pew Thian ; Fan, Yong ; Gilmore, John H. ; Lin, Weili ; Shen, Dinggang. / Neonatal brain MRI segmentation by building multi-region-multi-reference atlases. 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings. 2010. pp. 964-967
@inproceedings{d69f6bba6c2243508210c90bcc897b06,
title = "Neonatal brain MRI segmentation by building multi-region-multi-reference atlases",
abstract = "Neonatal brain MRI segmentation is challenging due to the poor image quality. Existing population atlases used for guiding segmentation are usually constructed by averaging all images in a population with no preference. However, such approaches diminish the important local inter-subject structural variability. In this paper, we propose a multiregion- multi-reference strategy for atlas building from a population. In brief, the brain is first parcellated into multiple anatomical regions, and for each region, the population images are classified into different subpopulations. The exemplars in sub-populations serve as structural references when determining the most suitable regional atlas for a to-be-segmented image. A final atlas is generated by combining all selected regional atlases, and a joint registration-segmentation strategy is employed for tissue segmentation. Experimental results demonstrate that segmentation with our atlas achieves high average tissue overlap rates with manual golden standard of 0.86 (SD 0.02) for gray matter (GM) and 0.83 (SD 0.03) for white matter (WM), and outperforms other atlases in comparison.",
keywords = "Joint registration-segmentation, Multiple atlases, Neonatal imaging, Tissue segmentation",
author = "Feng Shi and Yap, {Pew Thian} and Yong Fan and Gilmore, {John H.} and Weili Lin and Dinggang Shen",
year = "2010",
month = "8",
day = "9",
doi = "10.1109/ISBI.2010.5490148",
language = "English",
isbn = "9781424441266",
pages = "964--967",
booktitle = "2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings",

}

TY - GEN

T1 - Neonatal brain MRI segmentation by building multi-region-multi-reference atlases

AU - Shi, Feng

AU - Yap, Pew Thian

AU - Fan, Yong

AU - Gilmore, John H.

AU - Lin, Weili

AU - Shen, Dinggang

PY - 2010/8/9

Y1 - 2010/8/9

N2 - Neonatal brain MRI segmentation is challenging due to the poor image quality. Existing population atlases used for guiding segmentation are usually constructed by averaging all images in a population with no preference. However, such approaches diminish the important local inter-subject structural variability. In this paper, we propose a multiregion- multi-reference strategy for atlas building from a population. In brief, the brain is first parcellated into multiple anatomical regions, and for each region, the population images are classified into different subpopulations. The exemplars in sub-populations serve as structural references when determining the most suitable regional atlas for a to-be-segmented image. A final atlas is generated by combining all selected regional atlases, and a joint registration-segmentation strategy is employed for tissue segmentation. Experimental results demonstrate that segmentation with our atlas achieves high average tissue overlap rates with manual golden standard of 0.86 (SD 0.02) for gray matter (GM) and 0.83 (SD 0.03) for white matter (WM), and outperforms other atlases in comparison.

AB - Neonatal brain MRI segmentation is challenging due to the poor image quality. Existing population atlases used for guiding segmentation are usually constructed by averaging all images in a population with no preference. However, such approaches diminish the important local inter-subject structural variability. In this paper, we propose a multiregion- multi-reference strategy for atlas building from a population. In brief, the brain is first parcellated into multiple anatomical regions, and for each region, the population images are classified into different subpopulations. The exemplars in sub-populations serve as structural references when determining the most suitable regional atlas for a to-be-segmented image. A final atlas is generated by combining all selected regional atlases, and a joint registration-segmentation strategy is employed for tissue segmentation. Experimental results demonstrate that segmentation with our atlas achieves high average tissue overlap rates with manual golden standard of 0.86 (SD 0.02) for gray matter (GM) and 0.83 (SD 0.03) for white matter (WM), and outperforms other atlases in comparison.

KW - Joint registration-segmentation

KW - Multiple atlases

KW - Neonatal imaging

KW - Tissue segmentation

UR - http://www.scopus.com/inward/record.url?scp=77955205263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955205263&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2010.5490148

DO - 10.1109/ISBI.2010.5490148

M3 - Conference contribution

AN - SCOPUS:77955205263

SN - 9781424441266

SP - 964

EP - 967

BT - 2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings

ER -