Multiscale medial shape-based analysis of image objects

Stephen M. Pizer, Guido Gerig, Sarang Joshi, Stephen R. Aylward

    Research output: Contribution to journalArticle

    Abstract

    Medial representation of a three-dimensional (3-D) object or an ensemble of 3-D objects involves capturing the object interior as a locus of medial atoms, each atom being two vectors of equal length joined at the tail at the medial point. Medial representation has a variety of beneficial properties, among the most important of which are 1) its inherent geometry, provides an object-intrinsic coordinate system and thus provides correspondence between instances of the object in and near the object(s); 2) it captures the object interior and is, thus, very suitable for deformation; and 3) it provides the basis for an intuitive object-based multiscale sequence leading to efficiency of segmentation algorithms and trainability of statistical characterizations with limited training sets. As a result of these properties, medial representation is particularly suitable for the following image analysis tasks; how each operates will be described and will be illustrated by results: 1) segmentation of objects and object complexes via deformable models; 2) segmentation of tubular trees, e.g., of blood vessels, by following height ridges of measures of fit of medial atoms to target images; 3) object-based image registration via medial loci of such blood vessel trees; 4) statistical characterization of shape differences between control and pathological classes of structures. These analysis tasks are made possible by a new form of medial representation called m-reps, which is described.

    Original languageEnglish (US)
    Pages (from-to)1670-1679
    Number of pages10
    JournalProceedings of the IEEE
    Volume91
    Issue number10
    DOIs
    StatePublished - Oct 2003

    Fingerprint

    Blood vessels
    Atoms
    Image registration
    Image analysis
    Geometry

    Keywords

    • Discrimination
    • Medial
    • Medical image
    • Multiscale
    • Object
    • Registration
    • Segmentation
    • Shape
    • Statistics

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering

    Cite this

    Pizer, S. M., Gerig, G., Joshi, S., & Aylward, S. R. (2003). Multiscale medial shape-based analysis of image objects. Proceedings of the IEEE, 91(10), 1670-1679. https://doi.org/10.1109/JPROC.2003.817876

    Multiscale medial shape-based analysis of image objects. / Pizer, Stephen M.; Gerig, Guido; Joshi, Sarang; Aylward, Stephen R.

    In: Proceedings of the IEEE, Vol. 91, No. 10, 10.2003, p. 1670-1679.

    Research output: Contribution to journalArticle

    Pizer, SM, Gerig, G, Joshi, S & Aylward, SR 2003, 'Multiscale medial shape-based analysis of image objects', Proceedings of the IEEE, vol. 91, no. 10, pp. 1670-1679. https://doi.org/10.1109/JPROC.2003.817876
    Pizer, Stephen M. ; Gerig, Guido ; Joshi, Sarang ; Aylward, Stephen R. / Multiscale medial shape-based analysis of image objects. In: Proceedings of the IEEE. 2003 ; Vol. 91, No. 10. pp. 1670-1679.
    @article{555fde450af44b8eb01cc41807b984b5,
    title = "Multiscale medial shape-based analysis of image objects",
    abstract = "Medial representation of a three-dimensional (3-D) object or an ensemble of 3-D objects involves capturing the object interior as a locus of medial atoms, each atom being two vectors of equal length joined at the tail at the medial point. Medial representation has a variety of beneficial properties, among the most important of which are 1) its inherent geometry, provides an object-intrinsic coordinate system and thus provides correspondence between instances of the object in and near the object(s); 2) it captures the object interior and is, thus, very suitable for deformation; and 3) it provides the basis for an intuitive object-based multiscale sequence leading to efficiency of segmentation algorithms and trainability of statistical characterizations with limited training sets. As a result of these properties, medial representation is particularly suitable for the following image analysis tasks; how each operates will be described and will be illustrated by results: 1) segmentation of objects and object complexes via deformable models; 2) segmentation of tubular trees, e.g., of blood vessels, by following height ridges of measures of fit of medial atoms to target images; 3) object-based image registration via medial loci of such blood vessel trees; 4) statistical characterization of shape differences between control and pathological classes of structures. These analysis tasks are made possible by a new form of medial representation called m-reps, which is described.",
    keywords = "Discrimination, Medial, Medical image, Multiscale, Object, Registration, Segmentation, Shape, Statistics",
    author = "Pizer, {Stephen M.} and Guido Gerig and Sarang Joshi and Aylward, {Stephen R.}",
    year = "2003",
    month = "10",
    doi = "10.1109/JPROC.2003.817876",
    language = "English (US)",
    volume = "91",
    pages = "1670--1679",
    journal = "Proceedings of the IEEE",
    issn = "0018-9219",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "10",

    }

    TY - JOUR

    T1 - Multiscale medial shape-based analysis of image objects

    AU - Pizer, Stephen M.

    AU - Gerig, Guido

    AU - Joshi, Sarang

    AU - Aylward, Stephen R.

    PY - 2003/10

    Y1 - 2003/10

    N2 - Medial representation of a three-dimensional (3-D) object or an ensemble of 3-D objects involves capturing the object interior as a locus of medial atoms, each atom being two vectors of equal length joined at the tail at the medial point. Medial representation has a variety of beneficial properties, among the most important of which are 1) its inherent geometry, provides an object-intrinsic coordinate system and thus provides correspondence between instances of the object in and near the object(s); 2) it captures the object interior and is, thus, very suitable for deformation; and 3) it provides the basis for an intuitive object-based multiscale sequence leading to efficiency of segmentation algorithms and trainability of statistical characterizations with limited training sets. As a result of these properties, medial representation is particularly suitable for the following image analysis tasks; how each operates will be described and will be illustrated by results: 1) segmentation of objects and object complexes via deformable models; 2) segmentation of tubular trees, e.g., of blood vessels, by following height ridges of measures of fit of medial atoms to target images; 3) object-based image registration via medial loci of such blood vessel trees; 4) statistical characterization of shape differences between control and pathological classes of structures. These analysis tasks are made possible by a new form of medial representation called m-reps, which is described.

    AB - Medial representation of a three-dimensional (3-D) object or an ensemble of 3-D objects involves capturing the object interior as a locus of medial atoms, each atom being two vectors of equal length joined at the tail at the medial point. Medial representation has a variety of beneficial properties, among the most important of which are 1) its inherent geometry, provides an object-intrinsic coordinate system and thus provides correspondence between instances of the object in and near the object(s); 2) it captures the object interior and is, thus, very suitable for deformation; and 3) it provides the basis for an intuitive object-based multiscale sequence leading to efficiency of segmentation algorithms and trainability of statistical characterizations with limited training sets. As a result of these properties, medial representation is particularly suitable for the following image analysis tasks; how each operates will be described and will be illustrated by results: 1) segmentation of objects and object complexes via deformable models; 2) segmentation of tubular trees, e.g., of blood vessels, by following height ridges of measures of fit of medial atoms to target images; 3) object-based image registration via medial loci of such blood vessel trees; 4) statistical characterization of shape differences between control and pathological classes of structures. These analysis tasks are made possible by a new form of medial representation called m-reps, which is described.

    KW - Discrimination

    KW - Medial

    KW - Medical image

    KW - Multiscale

    KW - Object

    KW - Registration

    KW - Segmentation

    KW - Shape

    KW - Statistics

    UR - http://www.scopus.com/inward/record.url?scp=0242649130&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0242649130&partnerID=8YFLogxK

    U2 - 10.1109/JPROC.2003.817876

    DO - 10.1109/JPROC.2003.817876

    M3 - Article

    AN - SCOPUS:0242649130

    VL - 91

    SP - 1670

    EP - 1679

    JO - Proceedings of the IEEE

    JF - Proceedings of the IEEE

    SN - 0018-9219

    IS - 10

    ER -