Robust whole-brain segmentation: Application to traumatic brain injury

Christian Ledig (Corresponding Author), Rolf A. Heckemann, Alexander Hammers, Juan Carlos Lopez, Virginia F.J. Newcombe, Antonios Makropoulos, Jyrki Lötjönen, David K. Menon, Daniel Rueckert

Research output: Contribution to journalArticleScientificpeer-review

59 Citations (Scopus)

Abstract

We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.
Original languageEnglish
Pages (from-to)40-58
JournalMedical Image Analysis
Volume21
Issue number1
DOIs
Publication statusPublished - 2015
MoE publication typeA1 Journal article-refereed

Fingerprint

Atlases
Brain
Magnetic Resonance Spectroscopy
Labels
Magnetic resonance
Fusion reactions
Benchmarking
Globus Pallidus
Putamen
Brain Diseases
Thalamus
Disease Progression
Traumatic Brain Injury
Hippocampus
Anatomy
Joints
Biomarkers
Pathology
Labeling
Imaging techniques

Keywords

  • Traumatic brain injury
  • Magnetic resonance imaging
  • Multiatlas segmentation
  • Brain image segmentation
  • Expectation–maximisation

Cite this

Ledig, C., Heckemann, R. A., Hammers, A., Lopez, J. C., Newcombe, V. F. J., Makropoulos, A., ... Rueckert, D. (2015). Robust whole-brain segmentation: Application to traumatic brain injury. Medical Image Analysis, 21(1), 40-58. https://doi.org/10.1016/j.media.2014.12.003
Ledig, Christian ; Heckemann, Rolf A. ; Hammers, Alexander ; Lopez, Juan Carlos ; Newcombe, Virginia F.J. ; Makropoulos, Antonios ; Lötjönen, Jyrki ; Menon, David K. ; Rueckert, Daniel. / Robust whole-brain segmentation : Application to traumatic brain injury. In: Medical Image Analysis. 2015 ; Vol. 21, No. 1. pp. 40-58.
@article{6727ba0d40224a519454072e786481d2,
title = "Robust whole-brain segmentation: Application to traumatic brain injury",
abstract = "We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called {"}Multi-Atlas Label Propagation with Expectation-Maximisation based refinement{"} (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7{\%} accuracy using acute-phase MR images and 66.8{\%} accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0{\%} accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.",
keywords = "Traumatic brain injury, Magnetic resonance imaging, Multiatlas segmentation, Brain image segmentation, Expectation–maximisation",
author = "Christian Ledig and Heckemann, {Rolf A.} and Alexander Hammers and Lopez, {Juan Carlos} and Newcombe, {Virginia F.J.} and Antonios Makropoulos and Jyrki L{\"o}tj{\"o}nen and Menon, {David K.} and Daniel Rueckert",
year = "2015",
doi = "10.1016/j.media.2014.12.003",
language = "English",
volume = "21",
pages = "40--58",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "1",

}

Ledig, C, Heckemann, RA, Hammers, A, Lopez, JC, Newcombe, VFJ, Makropoulos, A, Lötjönen, J, Menon, DK & Rueckert, D 2015, 'Robust whole-brain segmentation: Application to traumatic brain injury', Medical Image Analysis, vol. 21, no. 1, pp. 40-58. https://doi.org/10.1016/j.media.2014.12.003

Robust whole-brain segmentation : Application to traumatic brain injury. / Ledig, Christian (Corresponding Author); Heckemann, Rolf A.; Hammers, Alexander; Lopez, Juan Carlos; Newcombe, Virginia F.J.; Makropoulos, Antonios; Lötjönen, Jyrki; Menon, David K.; Rueckert, Daniel.

In: Medical Image Analysis, Vol. 21, No. 1, 2015, p. 40-58.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Robust whole-brain segmentation

T2 - Application to traumatic brain injury

AU - Ledig, Christian

AU - Heckemann, Rolf A.

AU - Hammers, Alexander

AU - Lopez, Juan Carlos

AU - Newcombe, Virginia F.J.

AU - Makropoulos, Antonios

AU - Lötjönen, Jyrki

AU - Menon, David K.

AU - Rueckert, Daniel

PY - 2015

Y1 - 2015

N2 - We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.

AB - We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.

KW - Traumatic brain injury

KW - Magnetic resonance imaging

KW - Multiatlas segmentation

KW - Brain image segmentation

KW - Expectation–maximisation

U2 - 10.1016/j.media.2014.12.003

DO - 10.1016/j.media.2014.12.003

M3 - Article

VL - 21

SP - 40

EP - 58

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 1

ER -

Ledig C, Heckemann RA, Hammers A, Lopez JC, Newcombe VFJ, Makropoulos A et al. Robust whole-brain segmentation: Application to traumatic brain injury. Medical Image Analysis. 2015;21(1):40-58. https://doi.org/10.1016/j.media.2014.12.003