Statistical fusion of surface labels provided by multiple raters

John A. Bogovic, Bennett A. Landman, Pierre Louis Bazin, Jerry Ladd Prince

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Studies of the size and morphology of anatomical structures rely on accurate and reproducible delineation of the structures, obtained either by human raters or automatic segmentation algorithms. Measures of reproducibility and variability are vital aspects of such studies and are usually estimated using repeated scans or repeated delineations (in the case of human raters). Methods exist for simultaneously estimating the true structure and rater performance parameters from multiple segmentations and have been demonstrated on volumetric images. In this work, we extend the applicability of previous methods onto two-dimensional surfaces parameterized as triangle meshes. Label homogeneity is enforced using a Markov random field formulated with an energy that addresses the challenges introduced by the surface parameterization. The method was tested using both simulated raters and cortical gyral labels. Simulated raters are computed using a global error model as well as a novel and more realistic boundary error model. We study the impact of raters and their accuracy based on both models, and show how effectively this method estimates the true segmentation on simulated surfaces. The Markov random field formulation was shown to effectively enforce homogeneity for raters suffering from label noise. We demonstrated that our method provides substantial improvements in accuracy over single-atlas methods for all experimental conditions.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume7623
EditionPART 1
DOIs
StatePublished - 2010
EventMedical Imaging 2010: Image Processing - San Diego, CA, United States
Duration: Feb 14 2010Feb 16 2010

Other

OtherMedical Imaging 2010: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/14/102/16/10

Fingerprint

Labels
Fusion reactions
delineation
fusion
homogeneity
Parameterization
parameterization
triangles
mesh
estimating
formulations
Atlases
estimates
Noise
energy

Keywords

  • data fusion
  • delineation
  • labeling
  • parcellation
  • Rater evaluation
  • STAPLE
  • statistics
  • surface mesh

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Bogovic, J. A., Landman, B. A., Bazin, P. L., & Prince, J. L. (2010). Statistical fusion of surface labels provided by multiple raters. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (PART 1 ed., Vol. 7623). [762344] https://doi.org/10.1117/12.844214

Statistical fusion of surface labels provided by multiple raters. / Bogovic, John A.; Landman, Bennett A.; Bazin, Pierre Louis; Prince, Jerry Ladd.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7623 PART 1. ed. 2010. 762344.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bogovic, JA, Landman, BA, Bazin, PL & Prince, JL 2010, Statistical fusion of surface labels provided by multiple raters. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. PART 1 edn, vol. 7623, 762344, Medical Imaging 2010: Image Processing, San Diego, CA, United States, 2/14/10. https://doi.org/10.1117/12.844214
Bogovic JA, Landman BA, Bazin PL, Prince JL. Statistical fusion of surface labels provided by multiple raters. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. PART 1 ed. Vol. 7623. 2010. 762344 https://doi.org/10.1117/12.844214
Bogovic, John A. ; Landman, Bennett A. ; Bazin, Pierre Louis ; Prince, Jerry Ladd. / Statistical fusion of surface labels provided by multiple raters. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7623 PART 1. ed. 2010.
@inproceedings{e5cb0d9c66c544298cc428d4f72341e1,
title = "Statistical fusion of surface labels provided by multiple raters",
abstract = "Studies of the size and morphology of anatomical structures rely on accurate and reproducible delineation of the structures, obtained either by human raters or automatic segmentation algorithms. Measures of reproducibility and variability are vital aspects of such studies and are usually estimated using repeated scans or repeated delineations (in the case of human raters). Methods exist for simultaneously estimating the true structure and rater performance parameters from multiple segmentations and have been demonstrated on volumetric images. In this work, we extend the applicability of previous methods onto two-dimensional surfaces parameterized as triangle meshes. Label homogeneity is enforced using a Markov random field formulated with an energy that addresses the challenges introduced by the surface parameterization. The method was tested using both simulated raters and cortical gyral labels. Simulated raters are computed using a global error model as well as a novel and more realistic boundary error model. We study the impact of raters and their accuracy based on both models, and show how effectively this method estimates the true segmentation on simulated surfaces. The Markov random field formulation was shown to effectively enforce homogeneity for raters suffering from label noise. We demonstrated that our method provides substantial improvements in accuracy over single-atlas methods for all experimental conditions.",
keywords = "data fusion, delineation, labeling, parcellation, Rater evaluation, STAPLE, statistics, surface mesh",
author = "Bogovic, {John A.} and Landman, {Bennett A.} and Bazin, {Pierre Louis} and Prince, {Jerry Ladd}",
year = "2010",
doi = "10.1117/12.844214",
language = "English (US)",
isbn = "9780819480248",
volume = "7623",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
edition = "PART 1",

}

TY - GEN

T1 - Statistical fusion of surface labels provided by multiple raters

AU - Bogovic, John A.

AU - Landman, Bennett A.

AU - Bazin, Pierre Louis

AU - Prince, Jerry Ladd

PY - 2010

Y1 - 2010

N2 - Studies of the size and morphology of anatomical structures rely on accurate and reproducible delineation of the structures, obtained either by human raters or automatic segmentation algorithms. Measures of reproducibility and variability are vital aspects of such studies and are usually estimated using repeated scans or repeated delineations (in the case of human raters). Methods exist for simultaneously estimating the true structure and rater performance parameters from multiple segmentations and have been demonstrated on volumetric images. In this work, we extend the applicability of previous methods onto two-dimensional surfaces parameterized as triangle meshes. Label homogeneity is enforced using a Markov random field formulated with an energy that addresses the challenges introduced by the surface parameterization. The method was tested using both simulated raters and cortical gyral labels. Simulated raters are computed using a global error model as well as a novel and more realistic boundary error model. We study the impact of raters and their accuracy based on both models, and show how effectively this method estimates the true segmentation on simulated surfaces. The Markov random field formulation was shown to effectively enforce homogeneity for raters suffering from label noise. We demonstrated that our method provides substantial improvements in accuracy over single-atlas methods for all experimental conditions.

AB - Studies of the size and morphology of anatomical structures rely on accurate and reproducible delineation of the structures, obtained either by human raters or automatic segmentation algorithms. Measures of reproducibility and variability are vital aspects of such studies and are usually estimated using repeated scans or repeated delineations (in the case of human raters). Methods exist for simultaneously estimating the true structure and rater performance parameters from multiple segmentations and have been demonstrated on volumetric images. In this work, we extend the applicability of previous methods onto two-dimensional surfaces parameterized as triangle meshes. Label homogeneity is enforced using a Markov random field formulated with an energy that addresses the challenges introduced by the surface parameterization. The method was tested using both simulated raters and cortical gyral labels. Simulated raters are computed using a global error model as well as a novel and more realistic boundary error model. We study the impact of raters and their accuracy based on both models, and show how effectively this method estimates the true segmentation on simulated surfaces. The Markov random field formulation was shown to effectively enforce homogeneity for raters suffering from label noise. We demonstrated that our method provides substantial improvements in accuracy over single-atlas methods for all experimental conditions.

KW - data fusion

KW - delineation

KW - labeling

KW - parcellation

KW - Rater evaluation

KW - STAPLE

KW - statistics

KW - surface mesh

UR - http://www.scopus.com/inward/record.url?scp=79751511023&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79751511023&partnerID=8YFLogxK

U2 - 10.1117/12.844214

DO - 10.1117/12.844214

M3 - Conference contribution

AN - SCOPUS:79751511023

SN - 9780819480248

VL - 7623

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -