Semi-supervised cluster analysis of imaging data

Roman Filipovych, Susan M. Resnick, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

In this paper, we present a semi-supervised clustering-based framework for discovering coherent subpopulations in heterogeneous image sets. Our approach involves limited supervision in the form of labeled instances from two distributions that reflect a rough guess about subspace of features that are relevant for cluster analysis. By assuming that images are defined in a common space via registration to a common template, we propose a segmentation-based method for detecting locations that signify local regional differences in the two labeled sets. A PCA model of local image appearance is then estimated at each location of interest, and ranked with respect to its relevance for clustering. We develop an incremental k-means-like algorithm that discovers novel meaningful categories in a test image set. The application of our approach in this paper is in analysis of populations of healthy older adults. We validate our approach on a synthetic dataset, as well as on a dataset of brain images of older adults. We assess our method's performance on the problem of discovering clusters of MR images of human brain, and present a cluster-based measure of pathology that reflects the deviation of a subject's MR image from normal (i.e. cognitively stable) state. We analyze the clusters' structure, and show that clustering results obtained using our approach correlate well with clinical data.

Original languageEnglish (US)
Pages (from-to)2185-2197
Number of pages13
JournalNeuroImage
Volume54
Issue number3
DOIs
StatePublished - Feb 1 2011
Externally publishedYes

Fingerprint

Cluster Analysis
Passive Cutaneous Anaphylaxis
Brain
Pathology
Population
Datasets

Keywords

  • Aging
  • Cluster analysis
  • MCI
  • MRI
  • Semi-supervised pattern analysis

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Filipovych, R., Resnick, S. M., & Davatzikos, C. (2011). Semi-supervised cluster analysis of imaging data. NeuroImage, 54(3), 2185-2197. https://doi.org/10.1016/j.neuroimage.2010.09.074

Semi-supervised cluster analysis of imaging data. / Filipovych, Roman; Resnick, Susan M.; Davatzikos, Christos.

In: NeuroImage, Vol. 54, No. 3, 01.02.2011, p. 2185-2197.

Research output: Contribution to journalArticle

Filipovych, R, Resnick, SM & Davatzikos, C 2011, 'Semi-supervised cluster analysis of imaging data', NeuroImage, vol. 54, no. 3, pp. 2185-2197. https://doi.org/10.1016/j.neuroimage.2010.09.074
Filipovych R, Resnick SM, Davatzikos C. Semi-supervised cluster analysis of imaging data. NeuroImage. 2011 Feb 1;54(3):2185-2197. https://doi.org/10.1016/j.neuroimage.2010.09.074
Filipovych, Roman ; Resnick, Susan M. ; Davatzikos, Christos. / Semi-supervised cluster analysis of imaging data. In: NeuroImage. 2011 ; Vol. 54, No. 3. pp. 2185-2197.
@article{0291ffeb3b53449d9920061e3ea64647,
title = "Semi-supervised cluster analysis of imaging data",
abstract = "In this paper, we present a semi-supervised clustering-based framework for discovering coherent subpopulations in heterogeneous image sets. Our approach involves limited supervision in the form of labeled instances from two distributions that reflect a rough guess about subspace of features that are relevant for cluster analysis. By assuming that images are defined in a common space via registration to a common template, we propose a segmentation-based method for detecting locations that signify local regional differences in the two labeled sets. A PCA model of local image appearance is then estimated at each location of interest, and ranked with respect to its relevance for clustering. We develop an incremental k-means-like algorithm that discovers novel meaningful categories in a test image set. The application of our approach in this paper is in analysis of populations of healthy older adults. We validate our approach on a synthetic dataset, as well as on a dataset of brain images of older adults. We assess our method's performance on the problem of discovering clusters of MR images of human brain, and present a cluster-based measure of pathology that reflects the deviation of a subject's MR image from normal (i.e. cognitively stable) state. We analyze the clusters' structure, and show that clustering results obtained using our approach correlate well with clinical data.",
keywords = "Aging, Cluster analysis, MCI, MRI, Semi-supervised pattern analysis",
author = "Roman Filipovych and Resnick, {Susan M.} and Christos Davatzikos",
year = "2011",
month = "2",
day = "1",
doi = "10.1016/j.neuroimage.2010.09.074",
language = "English (US)",
volume = "54",
pages = "2185--2197",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - Semi-supervised cluster analysis of imaging data

AU - Filipovych, Roman

AU - Resnick, Susan M.

AU - Davatzikos, Christos

PY - 2011/2/1

Y1 - 2011/2/1

N2 - In this paper, we present a semi-supervised clustering-based framework for discovering coherent subpopulations in heterogeneous image sets. Our approach involves limited supervision in the form of labeled instances from two distributions that reflect a rough guess about subspace of features that are relevant for cluster analysis. By assuming that images are defined in a common space via registration to a common template, we propose a segmentation-based method for detecting locations that signify local regional differences in the two labeled sets. A PCA model of local image appearance is then estimated at each location of interest, and ranked with respect to its relevance for clustering. We develop an incremental k-means-like algorithm that discovers novel meaningful categories in a test image set. The application of our approach in this paper is in analysis of populations of healthy older adults. We validate our approach on a synthetic dataset, as well as on a dataset of brain images of older adults. We assess our method's performance on the problem of discovering clusters of MR images of human brain, and present a cluster-based measure of pathology that reflects the deviation of a subject's MR image from normal (i.e. cognitively stable) state. We analyze the clusters' structure, and show that clustering results obtained using our approach correlate well with clinical data.

AB - In this paper, we present a semi-supervised clustering-based framework for discovering coherent subpopulations in heterogeneous image sets. Our approach involves limited supervision in the form of labeled instances from two distributions that reflect a rough guess about subspace of features that are relevant for cluster analysis. By assuming that images are defined in a common space via registration to a common template, we propose a segmentation-based method for detecting locations that signify local regional differences in the two labeled sets. A PCA model of local image appearance is then estimated at each location of interest, and ranked with respect to its relevance for clustering. We develop an incremental k-means-like algorithm that discovers novel meaningful categories in a test image set. The application of our approach in this paper is in analysis of populations of healthy older adults. We validate our approach on a synthetic dataset, as well as on a dataset of brain images of older adults. We assess our method's performance on the problem of discovering clusters of MR images of human brain, and present a cluster-based measure of pathology that reflects the deviation of a subject's MR image from normal (i.e. cognitively stable) state. We analyze the clusters' structure, and show that clustering results obtained using our approach correlate well with clinical data.

KW - Aging

KW - Cluster analysis

KW - MCI

KW - MRI

KW - Semi-supervised pattern analysis

UR - http://www.scopus.com/inward/record.url?scp=78650206305&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650206305&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2010.09.074

DO - 10.1016/j.neuroimage.2010.09.074

M3 - Article

C2 - 20933091

AN - SCOPUS:78650206305

VL - 54

SP - 2185

EP - 2197

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -