Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease

Zhennan Yan, Shaoting Zhang, Xiaofeng Liu, Dimitris N. Metaxas, Albert Montillo

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

14 Citations (Scopus)

Abstract

Accurate segmentation of the 30+ subcortical structures in MR images of whole diseased brains is challenging due to inter-subject variability and complex geometry of brain anatomy. However a clinically viable solution yielding precise segmentation of the structures would enable: 1) accurate, objective measurement of structure volumes many of which are associated with diseases such as Alzheimer's, 2) therapy monitoring and 3) drug development. Our contributions are two-fold. First we construct an extended adaptive statistical atlas method (EASA) to use a non-stationary relaxation factor rather than a global one. This permits finer control over adaptivity allowing 34 structures to be simultaneously segmented rather than just 4 as in [13]. Second we use the output of a weighted majority voting (WMV) label fusion multi-atlas method as the input to EASA in a hybrid WMV-EASA approach. We assess our proposed approaches on 18 healthy subjects in the public IBSR database and on 9 subjects with Alzheimer's disease in the AIBL database. EASA is shown to produce state-of-the-art accuracy on healthy brains in a fraction of the time of comparable methods, while our hybrid WMV-EASA visibly improves segmentation accuracy for structures throughout the diseased brains.

Original languageEnglish (US)
Title of host publicationISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging
Subtitle of host publicationFrom Nano to Macro
Pages1202-1205
Number of pages4
DOIs
StatePublished - Aug 22 2013
Event2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 - San Francisco, CA, United States
Duration: Apr 7 2013Apr 11 2013

Other

Other2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013
CountryUnited States
CitySan Francisco, CA
Period4/7/134/11/13

Fingerprint

Atlases
Brain
Statistical methods
Alzheimer Disease
Politics
Brain Diseases
Databases
Labels
Drug Monitoring
Fusion reactions
Anatomy
Healthy Volunteers
Geometry
Monitoring

Keywords

  • Alzheimer's
  • brain segmentation
  • Dirichlet distribution
  • EM
  • label fusion
  • MRF
  • statistical atlas

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Yan, Z., Zhang, S., Liu, X., Metaxas, D. N., & Montillo, A. (2013). Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease. In ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro (pp. 1202-1205). [6556696] https://doi.org/10.1109/ISBI.2013.6556696

Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease. / Yan, Zhennan; Zhang, Shaoting; Liu, Xiaofeng; Metaxas, Dimitris N.; Montillo, Albert.

ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. p. 1202-1205 6556696.

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

Yan, Z, Zhang, S, Liu, X, Metaxas, DN & Montillo, A 2013, Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease. in ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro., 6556696, pp. 1202-1205, 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013, San Francisco, CA, United States, 4/7/13. https://doi.org/10.1109/ISBI.2013.6556696
Yan Z, Zhang S, Liu X, Metaxas DN, Montillo A. Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease. In ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. p. 1202-1205. 6556696 https://doi.org/10.1109/ISBI.2013.6556696
Yan, Zhennan ; Zhang, Shaoting ; Liu, Xiaofeng ; Metaxas, Dimitris N. ; Montillo, Albert. / Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease. ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. pp. 1202-1205
@inproceedings{0026eb2b4050401092a95caf5fbf8aff,
title = "Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease",
abstract = "Accurate segmentation of the 30+ subcortical structures in MR images of whole diseased brains is challenging due to inter-subject variability and complex geometry of brain anatomy. However a clinically viable solution yielding precise segmentation of the structures would enable: 1) accurate, objective measurement of structure volumes many of which are associated with diseases such as Alzheimer's, 2) therapy monitoring and 3) drug development. Our contributions are two-fold. First we construct an extended adaptive statistical atlas method (EASA) to use a non-stationary relaxation factor rather than a global one. This permits finer control over adaptivity allowing 34 structures to be simultaneously segmented rather than just 4 as in [13]. Second we use the output of a weighted majority voting (WMV) label fusion multi-atlas method as the input to EASA in a hybrid WMV-EASA approach. We assess our proposed approaches on 18 healthy subjects in the public IBSR database and on 9 subjects with Alzheimer's disease in the AIBL database. EASA is shown to produce state-of-the-art accuracy on healthy brains in a fraction of the time of comparable methods, while our hybrid WMV-EASA visibly improves segmentation accuracy for structures throughout the diseased brains.",
keywords = "Alzheimer's, brain segmentation, Dirichlet distribution, EM, label fusion, MRF, statistical atlas",
author = "Zhennan Yan and Shaoting Zhang and Xiaofeng Liu and Metaxas, {Dimitris N.} and Albert Montillo",
year = "2013",
month = "8",
day = "22",
doi = "10.1109/ISBI.2013.6556696",
language = "English (US)",
isbn = "9781467364546",
pages = "1202--1205",
booktitle = "ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Accurate segmentation of brain images into 34 structures combining a non-stationary adaptive statistical atlas and a multi-atlas with applications to Alzheimer'S disease

AU - Yan, Zhennan

AU - Zhang, Shaoting

AU - Liu, Xiaofeng

AU - Metaxas, Dimitris N.

AU - Montillo, Albert

PY - 2013/8/22

Y1 - 2013/8/22

N2 - Accurate segmentation of the 30+ subcortical structures in MR images of whole diseased brains is challenging due to inter-subject variability and complex geometry of brain anatomy. However a clinically viable solution yielding precise segmentation of the structures would enable: 1) accurate, objective measurement of structure volumes many of which are associated with diseases such as Alzheimer's, 2) therapy monitoring and 3) drug development. Our contributions are two-fold. First we construct an extended adaptive statistical atlas method (EASA) to use a non-stationary relaxation factor rather than a global one. This permits finer control over adaptivity allowing 34 structures to be simultaneously segmented rather than just 4 as in [13]. Second we use the output of a weighted majority voting (WMV) label fusion multi-atlas method as the input to EASA in a hybrid WMV-EASA approach. We assess our proposed approaches on 18 healthy subjects in the public IBSR database and on 9 subjects with Alzheimer's disease in the AIBL database. EASA is shown to produce state-of-the-art accuracy on healthy brains in a fraction of the time of comparable methods, while our hybrid WMV-EASA visibly improves segmentation accuracy for structures throughout the diseased brains.

AB - Accurate segmentation of the 30+ subcortical structures in MR images of whole diseased brains is challenging due to inter-subject variability and complex geometry of brain anatomy. However a clinically viable solution yielding precise segmentation of the structures would enable: 1) accurate, objective measurement of structure volumes many of which are associated with diseases such as Alzheimer's, 2) therapy monitoring and 3) drug development. Our contributions are two-fold. First we construct an extended adaptive statistical atlas method (EASA) to use a non-stationary relaxation factor rather than a global one. This permits finer control over adaptivity allowing 34 structures to be simultaneously segmented rather than just 4 as in [13]. Second we use the output of a weighted majority voting (WMV) label fusion multi-atlas method as the input to EASA in a hybrid WMV-EASA approach. We assess our proposed approaches on 18 healthy subjects in the public IBSR database and on 9 subjects with Alzheimer's disease in the AIBL database. EASA is shown to produce state-of-the-art accuracy on healthy brains in a fraction of the time of comparable methods, while our hybrid WMV-EASA visibly improves segmentation accuracy for structures throughout the diseased brains.

KW - Alzheimer's

KW - brain segmentation

KW - Dirichlet distribution

KW - EM

KW - label fusion

KW - MRF

KW - statistical atlas

UR - http://www.scopus.com/inward/record.url?scp=84881657655&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881657655&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2013.6556696

DO - 10.1109/ISBI.2013.6556696

M3 - Conference contribution

AN - SCOPUS:84881657655

SN - 9781467364546

SP - 1202

EP - 1205

BT - ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging

ER -