An analysis of methods for the selection of atlases for use in medical image segmentation

Jeffrey W. Prescott, Thomas Best, Furqan Haq, Rebecca Jackson, Metin Gurcan

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

Abstract

The use of atlases has been shown to be a robust method for segmentation of medical images. In this paper we explore different methods of selection of atlases for the segmentation of the quadriceps muscles in magnetic resonance (MR) images, although the results are pertinent for a wide range of applications. The experiments were performed using 103 images from the Osteoarthritis Initiative (OAI). The images were randomly split into a training set consisting of 50 images and a testing set of 53 images. Three different atlas selection methods were systematically compared. First, a set of readers was assigned the task of selecting atlases from a training population of images, which were selected to be representative subgroups of the total population. Second, the same readers were instructed to select atlases from a subset of the training data which was stratified based on population modes. Finally, every image in the training set was employed as an atlas, with no input from the readers, and the atlas which had the best initial registration, judged by an appropriate registration metric, was used in the final segmentation procedure. The segmentation results were quantified using the Zijdenbos similarity index (ZSI). The results show that over all readers the agreement of the segmentation algorithm decreased from 0.76 to 0.74 when using population modes to assist in atlas selection. The use of every image in the training set as an atlas outperformed both manual atlas selection methods, achieving a ZSI of 0.82.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2010
Subtitle of host publicationImage Processing
Volume7623
EditionPART 1
DOIs
StatePublished - Dec 1 2010
Externally publishedYes
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

Atlases
Magnetic resonance
Image segmentation
Muscle
Testing
readers
education
Experiments
Population
subgroups
muscles
Quadriceps Muscle
set theory
magnetic resonance
Osteoarthritis
Magnetic Resonance Spectroscopy

Keywords

  • atlases
  • osteoarthritis
  • segmentation

ASJC Scopus subject areas

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

Cite this

Prescott, J. W., Best, T., Haq, F., Jackson, R., & Gurcan, M. (2010). An analysis of methods for the selection of atlases for use in medical image segmentation. In Medical Imaging 2010: Image Processing (PART 1 ed., Vol. 7623). [76231T] https://doi.org/10.1117/12.843745

An analysis of methods for the selection of atlases for use in medical image segmentation. / Prescott, Jeffrey W.; Best, Thomas; Haq, Furqan; Jackson, Rebecca; Gurcan, Metin.

Medical Imaging 2010: Image Processing. Vol. 7623 PART 1. ed. 2010. 76231T.

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

Prescott, JW, Best, T, Haq, F, Jackson, R & Gurcan, M 2010, An analysis of methods for the selection of atlases for use in medical image segmentation. in Medical Imaging 2010: Image Processing. PART 1 edn, vol. 7623, 76231T, Medical Imaging 2010: Image Processing, San Diego, CA, United States, 2/14/10. https://doi.org/10.1117/12.843745
Prescott JW, Best T, Haq F, Jackson R, Gurcan M. An analysis of methods for the selection of atlases for use in medical image segmentation. In Medical Imaging 2010: Image Processing. PART 1 ed. Vol. 7623. 2010. 76231T https://doi.org/10.1117/12.843745
Prescott, Jeffrey W. ; Best, Thomas ; Haq, Furqan ; Jackson, Rebecca ; Gurcan, Metin. / An analysis of methods for the selection of atlases for use in medical image segmentation. Medical Imaging 2010: Image Processing. Vol. 7623 PART 1. ed. 2010.
@inproceedings{ac52da00af7a4ad8ace75cde236ad146,
title = "An analysis of methods for the selection of atlases for use in medical image segmentation",
abstract = "The use of atlases has been shown to be a robust method for segmentation of medical images. In this paper we explore different methods of selection of atlases for the segmentation of the quadriceps muscles in magnetic resonance (MR) images, although the results are pertinent for a wide range of applications. The experiments were performed using 103 images from the Osteoarthritis Initiative (OAI). The images were randomly split into a training set consisting of 50 images and a testing set of 53 images. Three different atlas selection methods were systematically compared. First, a set of readers was assigned the task of selecting atlases from a training population of images, which were selected to be representative subgroups of the total population. Second, the same readers were instructed to select atlases from a subset of the training data which was stratified based on population modes. Finally, every image in the training set was employed as an atlas, with no input from the readers, and the atlas which had the best initial registration, judged by an appropriate registration metric, was used in the final segmentation procedure. The segmentation results were quantified using the Zijdenbos similarity index (ZSI). The results show that over all readers the agreement of the segmentation algorithm decreased from 0.76 to 0.74 when using population modes to assist in atlas selection. The use of every image in the training set as an atlas outperformed both manual atlas selection methods, achieving a ZSI of 0.82.",
keywords = "atlases, osteoarthritis, segmentation",
author = "Prescott, {Jeffrey W.} and Thomas Best and Furqan Haq and Rebecca Jackson and Metin Gurcan",
year = "2010",
month = "12",
day = "1",
doi = "10.1117/12.843745",
language = "English (US)",
isbn = "9780819480248",
volume = "7623",
booktitle = "Medical Imaging 2010",
edition = "PART 1",

}

TY - GEN

T1 - An analysis of methods for the selection of atlases for use in medical image segmentation

AU - Prescott, Jeffrey W.

AU - Best, Thomas

AU - Haq, Furqan

AU - Jackson, Rebecca

AU - Gurcan, Metin

PY - 2010/12/1

Y1 - 2010/12/1

N2 - The use of atlases has been shown to be a robust method for segmentation of medical images. In this paper we explore different methods of selection of atlases for the segmentation of the quadriceps muscles in magnetic resonance (MR) images, although the results are pertinent for a wide range of applications. The experiments were performed using 103 images from the Osteoarthritis Initiative (OAI). The images were randomly split into a training set consisting of 50 images and a testing set of 53 images. Three different atlas selection methods were systematically compared. First, a set of readers was assigned the task of selecting atlases from a training population of images, which were selected to be representative subgroups of the total population. Second, the same readers were instructed to select atlases from a subset of the training data which was stratified based on population modes. Finally, every image in the training set was employed as an atlas, with no input from the readers, and the atlas which had the best initial registration, judged by an appropriate registration metric, was used in the final segmentation procedure. The segmentation results were quantified using the Zijdenbos similarity index (ZSI). The results show that over all readers the agreement of the segmentation algorithm decreased from 0.76 to 0.74 when using population modes to assist in atlas selection. The use of every image in the training set as an atlas outperformed both manual atlas selection methods, achieving a ZSI of 0.82.

AB - The use of atlases has been shown to be a robust method for segmentation of medical images. In this paper we explore different methods of selection of atlases for the segmentation of the quadriceps muscles in magnetic resonance (MR) images, although the results are pertinent for a wide range of applications. The experiments were performed using 103 images from the Osteoarthritis Initiative (OAI). The images were randomly split into a training set consisting of 50 images and a testing set of 53 images. Three different atlas selection methods were systematically compared. First, a set of readers was assigned the task of selecting atlases from a training population of images, which were selected to be representative subgroups of the total population. Second, the same readers were instructed to select atlases from a subset of the training data which was stratified based on population modes. Finally, every image in the training set was employed as an atlas, with no input from the readers, and the atlas which had the best initial registration, judged by an appropriate registration metric, was used in the final segmentation procedure. The segmentation results were quantified using the Zijdenbos similarity index (ZSI). The results show that over all readers the agreement of the segmentation algorithm decreased from 0.76 to 0.74 when using population modes to assist in atlas selection. The use of every image in the training set as an atlas outperformed both manual atlas selection methods, achieving a ZSI of 0.82.

KW - atlases

KW - osteoarthritis

KW - segmentation

UR - http://www.scopus.com/inward/record.url?scp=79751500127&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79751500127&partnerID=8YFLogxK

U2 - 10.1117/12.843745

DO - 10.1117/12.843745

M3 - Conference contribution

SN - 9780819480248

VL - 7623

BT - Medical Imaging 2010

ER -