Extracting evolving pathologies via spectral clustering.

Elena Bernardis, Kilian M. Pohl, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

A bottleneck in the analysis of longitudinal MR scans with white matter brain lesions is the temporally consistent segmentation of the pathology. We identify pathologies in 3D+t(ime) within a spectral graph clustering framework. Our clustering approach simultaneously segments and tracks the evolving lesions by identifying characteristic image patterns at each time-point and voxel correspondences across time-points. For each 3D image, our method constructs a graph where weights between nodes capture the likeliness of two voxels belonging to the same region. Based on these weights, we then establish rough correspondences between graph nodes at different time-points along estimated pathology evolution directions. We combine the graphs by aligning the weights to a reference time-point, thus integrating temporal information across the 3D images, and formulate the 3D+t segmentation problem as a binary partitioning of this graph. The resulting segmentation is very robust to local intensity fluctuations and yields better results than segmentations generated for each time-point.

Original languageEnglish (US)
Pages (from-to)680-691
Number of pages12
JournalInformation processing in medical imaging : proceedings of the ... conference
Volume23
StatePublished - 2013
Externally publishedYes

Fingerprint

Cluster Analysis
Pathology
Weights and Measures
Brain

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Extracting evolving pathologies via spectral clustering. / Bernardis, Elena; Pohl, Kilian M.; Davatzikos, Christos.

In: Information processing in medical imaging : proceedings of the ... conference, Vol. 23, 2013, p. 680-691.

Research output: Contribution to journalArticle

Bernardis, Elena ; Pohl, Kilian M. ; Davatzikos, Christos. / Extracting evolving pathologies via spectral clustering. In: Information processing in medical imaging : proceedings of the ... conference. 2013 ; Vol. 23. pp. 680-691.
@article{9218e539d47a4577a9d9c0ae549909ee,
title = "Extracting evolving pathologies via spectral clustering.",
abstract = "A bottleneck in the analysis of longitudinal MR scans with white matter brain lesions is the temporally consistent segmentation of the pathology. We identify pathologies in 3D+t(ime) within a spectral graph clustering framework. Our clustering approach simultaneously segments and tracks the evolving lesions by identifying characteristic image patterns at each time-point and voxel correspondences across time-points. For each 3D image, our method constructs a graph where weights between nodes capture the likeliness of two voxels belonging to the same region. Based on these weights, we then establish rough correspondences between graph nodes at different time-points along estimated pathology evolution directions. We combine the graphs by aligning the weights to a reference time-point, thus integrating temporal information across the 3D images, and formulate the 3D+t segmentation problem as a binary partitioning of this graph. The resulting segmentation is very robust to local intensity fluctuations and yields better results than segmentations generated for each time-point.",
author = "Elena Bernardis and Pohl, {Kilian M.} and Christos Davatzikos",
year = "2013",
language = "English (US)",
volume = "23",
pages = "680--691",
journal = "Information processing in medical imaging : proceedings of the ... conference",
issn = "1011-2499",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Extracting evolving pathologies via spectral clustering.

AU - Bernardis, Elena

AU - Pohl, Kilian M.

AU - Davatzikos, Christos

PY - 2013

Y1 - 2013

N2 - A bottleneck in the analysis of longitudinal MR scans with white matter brain lesions is the temporally consistent segmentation of the pathology. We identify pathologies in 3D+t(ime) within a spectral graph clustering framework. Our clustering approach simultaneously segments and tracks the evolving lesions by identifying characteristic image patterns at each time-point and voxel correspondences across time-points. For each 3D image, our method constructs a graph where weights between nodes capture the likeliness of two voxels belonging to the same region. Based on these weights, we then establish rough correspondences between graph nodes at different time-points along estimated pathology evolution directions. We combine the graphs by aligning the weights to a reference time-point, thus integrating temporal information across the 3D images, and formulate the 3D+t segmentation problem as a binary partitioning of this graph. The resulting segmentation is very robust to local intensity fluctuations and yields better results than segmentations generated for each time-point.

AB - A bottleneck in the analysis of longitudinal MR scans with white matter brain lesions is the temporally consistent segmentation of the pathology. We identify pathologies in 3D+t(ime) within a spectral graph clustering framework. Our clustering approach simultaneously segments and tracks the evolving lesions by identifying characteristic image patterns at each time-point and voxel correspondences across time-points. For each 3D image, our method constructs a graph where weights between nodes capture the likeliness of two voxels belonging to the same region. Based on these weights, we then establish rough correspondences between graph nodes at different time-points along estimated pathology evolution directions. We combine the graphs by aligning the weights to a reference time-point, thus integrating temporal information across the 3D images, and formulate the 3D+t segmentation problem as a binary partitioning of this graph. The resulting segmentation is very robust to local intensity fluctuations and yields better results than segmentations generated for each time-point.

UR - http://www.scopus.com/inward/record.url?scp=84901255386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901255386&partnerID=8YFLogxK

M3 - Article

C2 - 24684009

AN - SCOPUS:84901255386

VL - 23

SP - 680

EP - 691

JO - Information processing in medical imaging : proceedings of the ... conference

JF - Information processing in medical imaging : proceedings of the ... conference

SN - 1011-2499

ER -