Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis

Research output: Contribution to journalArticleAcademicpeer-review

13 Citations (Scopus)
27 Downloads (Pure)

Abstract

In biomedical imaging reliable segmentation of objects (e.g., from small cells up to large organs) is of fundamental importance for automated medical diagnosis. New approaches for multiscale segmentation can considerably improve performance in case of natural variations in intensity, size, and shape. This paper aims at segmenting objects of interest based on shape contours and automatically finding multiple objects with different scales. The overall strategy of this work is to combine nonlinear segmentation with scales spaces and spectral decompositions. We generalize a variational segmentation model based on total variation using Bregman distances to construct an inverse scale space. This offers the new model to be accomplished by a scale analysis approach based on a spectral decomposition of the total variation. As a result we obtain a very efficient, (nearly) parameter-free multiscale segmentation method that comes with an adaptive regularization parameter choice. To address the variety of shapes and scales present in biomedical imaging we analyze synthetic cases clarifying the role of scale and the relationship of Wulff shapes and eigenfunctions. To underline the potential of our approach and to show its wide applicability we address three different experimental biomedical applications. In particular, we demonstrate the added benefit for identifying and classifying circulating tumor cells and present interesting results for network analysis in retina imaging. Due to the nature of underlying nonlinear diffusion, the mathematical concepts in this work offer promising extensions to nonlocal classification problems.
Original languageEnglish
Pages (from-to)111-146
Number of pages36
JournalSIAM journal on imaging sciences
Volume10
Issue number1
DOIs
Publication statusPublished - 8 Feb 2017

Fingerprint

Bregman Distance
Spectral Analysis
Nonlinear Analysis
Spectrum analysis
Segmentation
Imaging techniques
Biomedical Imaging
Spectral Decomposition
Scale Space
Decomposition
Electric network analysis
Eigenvalues and eigenfunctions
Wulff Shape
Tumors
Total Variation Distance
Cells
Nonlocal Problems
Biomedical Applications
Retina
Nonlinear Diffusion

Keywords

  • EWI-27772
  • Bregman iteration
  • MSC-65K10
  • MSC-35A15
  • Wulff shapes
  • eigenfunctions
  • multiscale segmentation
  • nonlinear spectral methods
  • circulating tumor cells
  • inverse scale space
  • total variation
  • METIS-318833
  • IR-103647
  • Chan-Vese method
  • MSC-68U10

Cite this

@article{65f315bdf1124d699e4bd148b7173cd6,
title = "Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis",
abstract = "In biomedical imaging reliable segmentation of objects (e.g., from small cells up to large organs) is of fundamental importance for automated medical diagnosis. New approaches for multiscale segmentation can considerably improve performance in case of natural variations in intensity, size, and shape. This paper aims at segmenting objects of interest based on shape contours and automatically finding multiple objects with different scales. The overall strategy of this work is to combine nonlinear segmentation with scales spaces and spectral decompositions. We generalize a variational segmentation model based on total variation using Bregman distances to construct an inverse scale space. This offers the new model to be accomplished by a scale analysis approach based on a spectral decomposition of the total variation. As a result we obtain a very efficient, (nearly) parameter-free multiscale segmentation method that comes with an adaptive regularization parameter choice. To address the variety of shapes and scales present in biomedical imaging we analyze synthetic cases clarifying the role of scale and the relationship of Wulff shapes and eigenfunctions. To underline the potential of our approach and to show its wide applicability we address three different experimental biomedical applications. In particular, we demonstrate the added benefit for identifying and classifying circulating tumor cells and present interesting results for network analysis in retina imaging. Due to the nature of underlying nonlinear diffusion, the mathematical concepts in this work offer promising extensions to nonlocal classification problems.",
keywords = "EWI-27772, Bregman iteration, MSC-65K10, MSC-35A15, Wulff shapes, eigenfunctions, multiscale segmentation, nonlinear spectral methods, circulating tumor cells, inverse scale space, total variation, METIS-318833, IR-103647, Chan-Vese method, MSC-68U10",
author = "Zeune, {Leonie Laura} and {van Dalum}, Guus and Terstappen, {Leonardus Wendelinus Mathias Marie} and {van Gils}, {Stephanus A.} and Christoph Brune",
year = "2017",
month = "2",
day = "8",
doi = "10.1137/16M1074503",
language = "English",
volume = "10",
pages = "111--146",
journal = "SIAM journal on imaging sciences",
issn = "1936-4954",
publisher = "Society for Industrial and Applied Mathematics Publications",
number = "1",

}

Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis. / Zeune, Leonie Laura; van Dalum, Guus; Terstappen, Leonardus Wendelinus Mathias Marie; van Gils, Stephanus A.; Brune, Christoph.

In: SIAM journal on imaging sciences, Vol. 10, No. 1, 08.02.2017, p. 111-146.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis

AU - Zeune, Leonie Laura

AU - van Dalum, Guus

AU - Terstappen, Leonardus Wendelinus Mathias Marie

AU - van Gils, Stephanus A.

AU - Brune, Christoph

PY - 2017/2/8

Y1 - 2017/2/8

N2 - In biomedical imaging reliable segmentation of objects (e.g., from small cells up to large organs) is of fundamental importance for automated medical diagnosis. New approaches for multiscale segmentation can considerably improve performance in case of natural variations in intensity, size, and shape. This paper aims at segmenting objects of interest based on shape contours and automatically finding multiple objects with different scales. The overall strategy of this work is to combine nonlinear segmentation with scales spaces and spectral decompositions. We generalize a variational segmentation model based on total variation using Bregman distances to construct an inverse scale space. This offers the new model to be accomplished by a scale analysis approach based on a spectral decomposition of the total variation. As a result we obtain a very efficient, (nearly) parameter-free multiscale segmentation method that comes with an adaptive regularization parameter choice. To address the variety of shapes and scales present in biomedical imaging we analyze synthetic cases clarifying the role of scale and the relationship of Wulff shapes and eigenfunctions. To underline the potential of our approach and to show its wide applicability we address three different experimental biomedical applications. In particular, we demonstrate the added benefit for identifying and classifying circulating tumor cells and present interesting results for network analysis in retina imaging. Due to the nature of underlying nonlinear diffusion, the mathematical concepts in this work offer promising extensions to nonlocal classification problems.

AB - In biomedical imaging reliable segmentation of objects (e.g., from small cells up to large organs) is of fundamental importance for automated medical diagnosis. New approaches for multiscale segmentation can considerably improve performance in case of natural variations in intensity, size, and shape. This paper aims at segmenting objects of interest based on shape contours and automatically finding multiple objects with different scales. The overall strategy of this work is to combine nonlinear segmentation with scales spaces and spectral decompositions. We generalize a variational segmentation model based on total variation using Bregman distances to construct an inverse scale space. This offers the new model to be accomplished by a scale analysis approach based on a spectral decomposition of the total variation. As a result we obtain a very efficient, (nearly) parameter-free multiscale segmentation method that comes with an adaptive regularization parameter choice. To address the variety of shapes and scales present in biomedical imaging we analyze synthetic cases clarifying the role of scale and the relationship of Wulff shapes and eigenfunctions. To underline the potential of our approach and to show its wide applicability we address three different experimental biomedical applications. In particular, we demonstrate the added benefit for identifying and classifying circulating tumor cells and present interesting results for network analysis in retina imaging. Due to the nature of underlying nonlinear diffusion, the mathematical concepts in this work offer promising extensions to nonlocal classification problems.

KW - EWI-27772

KW - Bregman iteration

KW - MSC-65K10

KW - MSC-35A15

KW - Wulff shapes

KW - eigenfunctions

KW - multiscale segmentation

KW - nonlinear spectral methods

KW - circulating tumor cells

KW - inverse scale space

KW - total variation

KW - METIS-318833

KW - IR-103647

KW - Chan-Vese method

KW - MSC-68U10

U2 - 10.1137/16M1074503

DO - 10.1137/16M1074503

M3 - Article

VL - 10

SP - 111

EP - 146

JO - SIAM journal on imaging sciences

JF - SIAM journal on imaging sciences

SN - 1936-4954

IS - 1

ER -