A Bayesian morphometry algorithm

Edward H. Herskovits, Hanchuan Peng, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

Most methods for structure-function analysis of the brain in medical images are usually based on voxel-wise statistical tests performed on registered magnetic resonance (MR) images across subjects. A major drawback of such methods is the inability to accurately locate regions that manifest nonlinear associations with clinical variables. In this paper, we propose Bayesian morphological analysis methods, based on a Bayesian-network representation, for the analysis of MR brain images. First, we describe how Bayesian networks (BNs) can represent probabilistic associations among voxels and clinical (function) variables. Second, we present a model-selection framework, which generates a BN that captures structure-function relationships from MR brain images and function variables. We demonstrate our methods in the context of determining associations between regional brain atrophy (as demonstrated on MR images of the brain), and functional deficits. We employ two data sets for this evaluation: the first contains MR images of 11 subjects, where associations between regional atrophy and a functional deficit are almost linear; the second data set contains MR images of the ventricles of 84 subjects, where the structure-function association is nonlinear. Our methods successfully identify voxel-wise morphological changes that are associated with functional deficits in both data sets, whereas standard statistical analysis (i.e., t-test and paired t-test) fails in the nonlinear-association case.

Original languageEnglish (US)
Pages (from-to)723-737
Number of pages15
JournalIEEE Transactions on Medical Imaging
Volume23
Issue number6
DOIs
StatePublished - Jun 2004
Externally publishedYes

Fingerprint

Magnetic resonance
Magnetic Resonance Spectroscopy
Brain
Bayesian networks
Atrophy
Bayes Theorem
Statistical tests
Statistical methods
Datasets

Keywords

  • Bayes procedures
  • Bayesian network
  • Computational anatomy
  • Image analysis
  • Image classification
  • Morphology-function analysis
  • Voxel-based morphometry

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Herskovits, E. H., Peng, H., & Davatzikos, C. (2004). A Bayesian morphometry algorithm. IEEE Transactions on Medical Imaging, 23(6), 723-737. https://doi.org/10.1109/TMI.2004.826949

A Bayesian morphometry algorithm. / Herskovits, Edward H.; Peng, Hanchuan; Davatzikos, Christos.

In: IEEE Transactions on Medical Imaging, Vol. 23, No. 6, 06.2004, p. 723-737.

Research output: Contribution to journalArticle

Herskovits, EH, Peng, H & Davatzikos, C 2004, 'A Bayesian morphometry algorithm', IEEE Transactions on Medical Imaging, vol. 23, no. 6, pp. 723-737. https://doi.org/10.1109/TMI.2004.826949
Herskovits, Edward H. ; Peng, Hanchuan ; Davatzikos, Christos. / A Bayesian morphometry algorithm. In: IEEE Transactions on Medical Imaging. 2004 ; Vol. 23, No. 6. pp. 723-737.
@article{0af0e238b2704d94b054a64d9b12d790,
title = "A Bayesian morphometry algorithm",
abstract = "Most methods for structure-function analysis of the brain in medical images are usually based on voxel-wise statistical tests performed on registered magnetic resonance (MR) images across subjects. A major drawback of such methods is the inability to accurately locate regions that manifest nonlinear associations with clinical variables. In this paper, we propose Bayesian morphological analysis methods, based on a Bayesian-network representation, for the analysis of MR brain images. First, we describe how Bayesian networks (BNs) can represent probabilistic associations among voxels and clinical (function) variables. Second, we present a model-selection framework, which generates a BN that captures structure-function relationships from MR brain images and function variables. We demonstrate our methods in the context of determining associations between regional brain atrophy (as demonstrated on MR images of the brain), and functional deficits. We employ two data sets for this evaluation: the first contains MR images of 11 subjects, where associations between regional atrophy and a functional deficit are almost linear; the second data set contains MR images of the ventricles of 84 subjects, where the structure-function association is nonlinear. Our methods successfully identify voxel-wise morphological changes that are associated with functional deficits in both data sets, whereas standard statistical analysis (i.e., t-test and paired t-test) fails in the nonlinear-association case.",
keywords = "Bayes procedures, Bayesian network, Computational anatomy, Image analysis, Image classification, Morphology-function analysis, Voxel-based morphometry",
author = "Herskovits, {Edward H.} and Hanchuan Peng and Christos Davatzikos",
year = "2004",
month = "6",
doi = "10.1109/TMI.2004.826949",
language = "English (US)",
volume = "23",
pages = "723--737",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

TY - JOUR

T1 - A Bayesian morphometry algorithm

AU - Herskovits, Edward H.

AU - Peng, Hanchuan

AU - Davatzikos, Christos

PY - 2004/6

Y1 - 2004/6

N2 - Most methods for structure-function analysis of the brain in medical images are usually based on voxel-wise statistical tests performed on registered magnetic resonance (MR) images across subjects. A major drawback of such methods is the inability to accurately locate regions that manifest nonlinear associations with clinical variables. In this paper, we propose Bayesian morphological analysis methods, based on a Bayesian-network representation, for the analysis of MR brain images. First, we describe how Bayesian networks (BNs) can represent probabilistic associations among voxels and clinical (function) variables. Second, we present a model-selection framework, which generates a BN that captures structure-function relationships from MR brain images and function variables. We demonstrate our methods in the context of determining associations between regional brain atrophy (as demonstrated on MR images of the brain), and functional deficits. We employ two data sets for this evaluation: the first contains MR images of 11 subjects, where associations between regional atrophy and a functional deficit are almost linear; the second data set contains MR images of the ventricles of 84 subjects, where the structure-function association is nonlinear. Our methods successfully identify voxel-wise morphological changes that are associated with functional deficits in both data sets, whereas standard statistical analysis (i.e., t-test and paired t-test) fails in the nonlinear-association case.

AB - Most methods for structure-function analysis of the brain in medical images are usually based on voxel-wise statistical tests performed on registered magnetic resonance (MR) images across subjects. A major drawback of such methods is the inability to accurately locate regions that manifest nonlinear associations with clinical variables. In this paper, we propose Bayesian morphological analysis methods, based on a Bayesian-network representation, for the analysis of MR brain images. First, we describe how Bayesian networks (BNs) can represent probabilistic associations among voxels and clinical (function) variables. Second, we present a model-selection framework, which generates a BN that captures structure-function relationships from MR brain images and function variables. We demonstrate our methods in the context of determining associations between regional brain atrophy (as demonstrated on MR images of the brain), and functional deficits. We employ two data sets for this evaluation: the first contains MR images of 11 subjects, where associations between regional atrophy and a functional deficit are almost linear; the second data set contains MR images of the ventricles of 84 subjects, where the structure-function association is nonlinear. Our methods successfully identify voxel-wise morphological changes that are associated with functional deficits in both data sets, whereas standard statistical analysis (i.e., t-test and paired t-test) fails in the nonlinear-association case.

KW - Bayes procedures

KW - Bayesian network

KW - Computational anatomy

KW - Image analysis

KW - Image classification

KW - Morphology-function analysis

KW - Voxel-based morphometry

UR - http://www.scopus.com/inward/record.url?scp=2942511605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942511605&partnerID=8YFLogxK

U2 - 10.1109/TMI.2004.826949

DO - 10.1109/TMI.2004.826949

M3 - Article

C2 - 15191147

AN - SCOPUS:2942511605

VL - 23

SP - 723

EP - 737

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 6

ER -