Bayesian clustering methods for morphological analysis of MR images

Hanckuan Peng, Edward Herskovits, Christos Davatzikos

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

Abstract

Determining the relationship between structure (i.e. morphology) and function is a fundamental problem in brain research. In this paper we present a new framework based on Bayesian clustering methods for the voxel-wise statistical morphology-function analysis of registered MR images. We construct a Bayesian network to automatically identify the significant associations between voxel-wise morphological variables and functional variables, such as cognitive performance. A Bayesian latent variable induction method is applied to locate the homogeneous association regions on registered maps of morphological variables. Experimental results on images with simulated atrophy confirm that the new method outperforms conventional statistical method, based on linear statistics.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
PublisherIEEE Computer Society
Pages875-878
Number of pages4
Volume2002-January
ISBN (Print)078037584X
DOIs
StatePublished - 2002
EventIEEE International Symposium on Biomedical Imaging, ISBI 2002 - Washington, United States
Duration: Jul 7 2002Jul 10 2002

Other

OtherIEEE International Symposium on Biomedical Imaging, ISBI 2002
CountryUnited States
CityWashington
Period7/7/027/10/02

Fingerprint

Bayes Theorem
Cluster Analysis
Bayesian networks
Brain
Statistical methods
Statistics
Atrophy
Research

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Peng, H., Herskovits, E., & Davatzikos, C. (2002). Bayesian clustering methods for morphological analysis of MR images. In Proceedings - International Symposium on Biomedical Imaging (Vol. 2002-January, pp. 875-878). [1029399] IEEE Computer Society. https://doi.org/10.1109/ISBI.2002.1029399

Bayesian clustering methods for morphological analysis of MR images. / Peng, Hanckuan; Herskovits, Edward; Davatzikos, Christos.

Proceedings - International Symposium on Biomedical Imaging. Vol. 2002-January IEEE Computer Society, 2002. p. 875-878 1029399.

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

Peng, H, Herskovits, E & Davatzikos, C 2002, Bayesian clustering methods for morphological analysis of MR images. in Proceedings - International Symposium on Biomedical Imaging. vol. 2002-January, 1029399, IEEE Computer Society, pp. 875-878, IEEE International Symposium on Biomedical Imaging, ISBI 2002, Washington, United States, 7/7/02. https://doi.org/10.1109/ISBI.2002.1029399
Peng H, Herskovits E, Davatzikos C. Bayesian clustering methods for morphological analysis of MR images. In Proceedings - International Symposium on Biomedical Imaging. Vol. 2002-January. IEEE Computer Society. 2002. p. 875-878. 1029399 https://doi.org/10.1109/ISBI.2002.1029399
Peng, Hanckuan ; Herskovits, Edward ; Davatzikos, Christos. / Bayesian clustering methods for morphological analysis of MR images. Proceedings - International Symposium on Biomedical Imaging. Vol. 2002-January IEEE Computer Society, 2002. pp. 875-878
@inproceedings{6f530272f8894f8fa38ac5fdc9537281,
title = "Bayesian clustering methods for morphological analysis of MR images",
abstract = "Determining the relationship between structure (i.e. morphology) and function is a fundamental problem in brain research. In this paper we present a new framework based on Bayesian clustering methods for the voxel-wise statistical morphology-function analysis of registered MR images. We construct a Bayesian network to automatically identify the significant associations between voxel-wise morphological variables and functional variables, such as cognitive performance. A Bayesian latent variable induction method is applied to locate the homogeneous association regions on registered maps of morphological variables. Experimental results on images with simulated atrophy confirm that the new method outperforms conventional statistical method, based on linear statistics.",
author = "Hanckuan Peng and Edward Herskovits and Christos Davatzikos",
year = "2002",
doi = "10.1109/ISBI.2002.1029399",
language = "English (US)",
isbn = "078037584X",
volume = "2002-January",
pages = "875--878",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Bayesian clustering methods for morphological analysis of MR images

AU - Peng, Hanckuan

AU - Herskovits, Edward

AU - Davatzikos, Christos

PY - 2002

Y1 - 2002

N2 - Determining the relationship between structure (i.e. morphology) and function is a fundamental problem in brain research. In this paper we present a new framework based on Bayesian clustering methods for the voxel-wise statistical morphology-function analysis of registered MR images. We construct a Bayesian network to automatically identify the significant associations between voxel-wise morphological variables and functional variables, such as cognitive performance. A Bayesian latent variable induction method is applied to locate the homogeneous association regions on registered maps of morphological variables. Experimental results on images with simulated atrophy confirm that the new method outperforms conventional statistical method, based on linear statistics.

AB - Determining the relationship between structure (i.e. morphology) and function is a fundamental problem in brain research. In this paper we present a new framework based on Bayesian clustering methods for the voxel-wise statistical morphology-function analysis of registered MR images. We construct a Bayesian network to automatically identify the significant associations between voxel-wise morphological variables and functional variables, such as cognitive performance. A Bayesian latent variable induction method is applied to locate the homogeneous association regions on registered maps of morphological variables. Experimental results on images with simulated atrophy confirm that the new method outperforms conventional statistical method, based on linear statistics.

UR - http://www.scopus.com/inward/record.url?scp=84948670256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948670256&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2002.1029399

DO - 10.1109/ISBI.2002.1029399

M3 - Conference contribution

SN - 078037584X

VL - 2002-January

SP - 875

EP - 878

BT - Proceedings - International Symposium on Biomedical Imaging

PB - IEEE Computer Society

ER -