Unsupervised tissue type segmentation of 3D dual-echo MR head data

Guido Gerig, John Martin, Ron Kikinis, Olaf Kubler, Martha Shenton, Ferenc A. Jolesz

    Research output: Contribution to journalArticle

    Abstract

    The visualization of 3D phenomena and the extraction of quantitative information from magnetic resonance (MR) image data require efficient semiautomated or automated segmentation techniques. The application of multivariate statistical classification to the segmentation of dual-echo volume data of the human head into tissue types (grey matter, white matter and fluid spaces) is studied in this paper. Tests of the radiometric variability of tissue classes within the data volume demonstrate the improvement of the image acquisition technology and the suitability of statistical methods to perform brain tissue segmentation. Supervised classification is successfully applied to a study of 16 MR volume images of the human head, illustrating the robustness of this method in segmenting brain (white and grey matter) and cerebrospinal fluid (CSF). To avoid subjective criteria involved in the supervised approach, ISODATA clustering as well as clustering based on nonparametric probability density estimation were tested. Both methods performed well (success rates 93.8% and 87.5%, respectively), indicating that the classification procedure can be completely automated. The reproducibility and reliability of supervised and unsupervised classfication were studied by comparing results of segmentation performed by five expert operators. Results suggest that the interoperator and intraoperator variations could be reduced using automated clustering techniques. The accuracy of the volume calculations was quantified by applying the MR imaging and segmentation process to a phantom resembling shape and tissue characteristics of brain tissue. The segmented brain objects are displayed using 3D surface rendering.

    Original languageEnglish (US)
    Pages (from-to)349-360
    Number of pages12
    JournalImage and Vision Computing
    Volume10
    Issue number6
    DOIs
    StatePublished - 1992

    Fingerprint

    Magnetic resonance
    Tissue
    Brain
    Cerebrospinal fluid
    Image acquisition
    Statistical methods
    Visualization
    Imaging techniques
    Fluids

    Keywords

    • 3D display
    • assessment of brain tissues
    • clustering
    • magnetic resonance volume data
    • medical image analysis
    • segmentation in multidimensions
    • statistical classification

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition
    • Signal Processing
    • Electrical and Electronic Engineering

    Cite this

    Gerig, G., Martin, J., Kikinis, R., Kubler, O., Shenton, M., & Jolesz, F. A. (1992). Unsupervised tissue type segmentation of 3D dual-echo MR head data. Image and Vision Computing, 10(6), 349-360. https://doi.org/10.1016/0262-8856(92)90021-T

    Unsupervised tissue type segmentation of 3D dual-echo MR head data. / Gerig, Guido; Martin, John; Kikinis, Ron; Kubler, Olaf; Shenton, Martha; Jolesz, Ferenc A.

    In: Image and Vision Computing, Vol. 10, No. 6, 1992, p. 349-360.

    Research output: Contribution to journalArticle

    Gerig, G, Martin, J, Kikinis, R, Kubler, O, Shenton, M & Jolesz, FA 1992, 'Unsupervised tissue type segmentation of 3D dual-echo MR head data', Image and Vision Computing, vol. 10, no. 6, pp. 349-360. https://doi.org/10.1016/0262-8856(92)90021-T
    Gerig, Guido ; Martin, John ; Kikinis, Ron ; Kubler, Olaf ; Shenton, Martha ; Jolesz, Ferenc A. / Unsupervised tissue type segmentation of 3D dual-echo MR head data. In: Image and Vision Computing. 1992 ; Vol. 10, No. 6. pp. 349-360.
    @article{865daf0f69484266a396e9c5ebbaa37f,
    title = "Unsupervised tissue type segmentation of 3D dual-echo MR head data",
    abstract = "The visualization of 3D phenomena and the extraction of quantitative information from magnetic resonance (MR) image data require efficient semiautomated or automated segmentation techniques. The application of multivariate statistical classification to the segmentation of dual-echo volume data of the human head into tissue types (grey matter, white matter and fluid spaces) is studied in this paper. Tests of the radiometric variability of tissue classes within the data volume demonstrate the improvement of the image acquisition technology and the suitability of statistical methods to perform brain tissue segmentation. Supervised classification is successfully applied to a study of 16 MR volume images of the human head, illustrating the robustness of this method in segmenting brain (white and grey matter) and cerebrospinal fluid (CSF). To avoid subjective criteria involved in the supervised approach, ISODATA clustering as well as clustering based on nonparametric probability density estimation were tested. Both methods performed well (success rates 93.8{\%} and 87.5{\%}, respectively), indicating that the classification procedure can be completely automated. The reproducibility and reliability of supervised and unsupervised classfication were studied by comparing results of segmentation performed by five expert operators. Results suggest that the interoperator and intraoperator variations could be reduced using automated clustering techniques. The accuracy of the volume calculations was quantified by applying the MR imaging and segmentation process to a phantom resembling shape and tissue characteristics of brain tissue. The segmented brain objects are displayed using 3D surface rendering.",
    keywords = "3D display, assessment of brain tissues, clustering, magnetic resonance volume data, medical image analysis, segmentation in multidimensions, statistical classification",
    author = "Guido Gerig and John Martin and Ron Kikinis and Olaf Kubler and Martha Shenton and Jolesz, {Ferenc A.}",
    year = "1992",
    doi = "10.1016/0262-8856(92)90021-T",
    language = "English (US)",
    volume = "10",
    pages = "349--360",
    journal = "Image and Vision Computing",
    issn = "0262-8856",
    publisher = "Elsevier Limited",
    number = "6",

    }

    TY - JOUR

    T1 - Unsupervised tissue type segmentation of 3D dual-echo MR head data

    AU - Gerig, Guido

    AU - Martin, John

    AU - Kikinis, Ron

    AU - Kubler, Olaf

    AU - Shenton, Martha

    AU - Jolesz, Ferenc A.

    PY - 1992

    Y1 - 1992

    N2 - The visualization of 3D phenomena and the extraction of quantitative information from magnetic resonance (MR) image data require efficient semiautomated or automated segmentation techniques. The application of multivariate statistical classification to the segmentation of dual-echo volume data of the human head into tissue types (grey matter, white matter and fluid spaces) is studied in this paper. Tests of the radiometric variability of tissue classes within the data volume demonstrate the improvement of the image acquisition technology and the suitability of statistical methods to perform brain tissue segmentation. Supervised classification is successfully applied to a study of 16 MR volume images of the human head, illustrating the robustness of this method in segmenting brain (white and grey matter) and cerebrospinal fluid (CSF). To avoid subjective criteria involved in the supervised approach, ISODATA clustering as well as clustering based on nonparametric probability density estimation were tested. Both methods performed well (success rates 93.8% and 87.5%, respectively), indicating that the classification procedure can be completely automated. The reproducibility and reliability of supervised and unsupervised classfication were studied by comparing results of segmentation performed by five expert operators. Results suggest that the interoperator and intraoperator variations could be reduced using automated clustering techniques. The accuracy of the volume calculations was quantified by applying the MR imaging and segmentation process to a phantom resembling shape and tissue characteristics of brain tissue. The segmented brain objects are displayed using 3D surface rendering.

    AB - The visualization of 3D phenomena and the extraction of quantitative information from magnetic resonance (MR) image data require efficient semiautomated or automated segmentation techniques. The application of multivariate statistical classification to the segmentation of dual-echo volume data of the human head into tissue types (grey matter, white matter and fluid spaces) is studied in this paper. Tests of the radiometric variability of tissue classes within the data volume demonstrate the improvement of the image acquisition technology and the suitability of statistical methods to perform brain tissue segmentation. Supervised classification is successfully applied to a study of 16 MR volume images of the human head, illustrating the robustness of this method in segmenting brain (white and grey matter) and cerebrospinal fluid (CSF). To avoid subjective criteria involved in the supervised approach, ISODATA clustering as well as clustering based on nonparametric probability density estimation were tested. Both methods performed well (success rates 93.8% and 87.5%, respectively), indicating that the classification procedure can be completely automated. The reproducibility and reliability of supervised and unsupervised classfication were studied by comparing results of segmentation performed by five expert operators. Results suggest that the interoperator and intraoperator variations could be reduced using automated clustering techniques. The accuracy of the volume calculations was quantified by applying the MR imaging and segmentation process to a phantom resembling shape and tissue characteristics of brain tissue. The segmented brain objects are displayed using 3D surface rendering.

    KW - 3D display

    KW - assessment of brain tissues

    KW - clustering

    KW - magnetic resonance volume data

    KW - medical image analysis

    KW - segmentation in multidimensions

    KW - statistical classification

    UR - http://www.scopus.com/inward/record.url?scp=0000515380&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0000515380&partnerID=8YFLogxK

    U2 - 10.1016/0262-8856(92)90021-T

    DO - 10.1016/0262-8856(92)90021-T

    M3 - Article

    VL - 10

    SP - 349

    EP - 360

    JO - Image and Vision Computing

    JF - Image and Vision Computing

    SN - 0262-8856

    IS - 6

    ER -