Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models

G. Székely, A. Kelemen, Ch Brechbühler, G. Gerig

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

    Abstract

    This paper describes a new model-based segmentation technique combining desirable properties of physical models (snakes, [2]), shape representation by Fourier parametrization (Fourier snakes, [12]), and modelling of natural shape variability (eigenmodes, [7, 10]). Flexible shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation with rcspect to a small sct of stable landmarks (ACPC in our application) and explaining the remaining variability among a series of images with the model flexibility. Although straightforward, the extension to 3-D is severely impeded by finding a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization [16, 17] which achieves a uniform mapping between object surface and parameter space. The 3D model building and Fourier-snake procedure are demonstrated by segmenting deep structures of the human brain from MR volume data.

    Original languageEnglish (US)
    Title of host publicationComputer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings
    PublisherSpringer Verlag
    Pages495-505
    Number of pages11
    Volume905
    ISBN (Print)9783540591207
    StatePublished - 1995
    Event1st International Conference on Computer Vision, Virtual Reality, and Robotics in Medicine, CVRMed 1995 - Nice, France
    Duration: Apr 3 1995Apr 6 1995

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume905
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other1st International Conference on Computer Vision, Virtual Reality, and Robotics in Medicine, CVRMed 1995
    CountryFrance
    CityNice
    Period4/3/954/6/95

    Fingerprint

    Elastic Deformation
    Snakes
    Elastic deformation
    Parametrization
    Magnetic resonance imaging
    Segmentation
    Shape Representation
    Landmarks
    Physical Model
    3D Model
    3D
    Parameter Space
    Flexibility
    Model
    Model-based
    Topology
    Series
    Arbitrary
    Brain
    Modeling

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Székely, G., Kelemen, A., Brechbühler, C., & Gerig, G. (1995). Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models. In Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings (Vol. 905, pp. 495-505). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 905). Springer Verlag.

    Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models. / Székely, G.; Kelemen, A.; Brechbühler, Ch; Gerig, G.

    Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings. Vol. 905 Springer Verlag, 1995. p. 495-505 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 905).

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

    Székely, G, Kelemen, A, Brechbühler, C & Gerig, G 1995, Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models. in Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings. vol. 905, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 905, Springer Verlag, pp. 495-505, 1st International Conference on Computer Vision, Virtual Reality, and Robotics in Medicine, CVRMed 1995, Nice, France, 4/3/95.
    Székely G, Kelemen A, Brechbühler C, Gerig G. Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models. In Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings. Vol. 905. Springer Verlag. 1995. p. 495-505. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    Székely, G. ; Kelemen, A. ; Brechbühler, Ch ; Gerig, G. / Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models. Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings. Vol. 905 Springer Verlag, 1995. pp. 495-505 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{e55650b40c1441028a629e031b0eceb9,
    title = "Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models",
    abstract = "This paper describes a new model-based segmentation technique combining desirable properties of physical models (snakes, [2]), shape representation by Fourier parametrization (Fourier snakes, [12]), and modelling of natural shape variability (eigenmodes, [7, 10]). Flexible shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation with rcspect to a small sct of stable landmarks (ACPC in our application) and explaining the remaining variability among a series of images with the model flexibility. Although straightforward, the extension to 3-D is severely impeded by finding a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization [16, 17] which achieves a uniform mapping between object surface and parameter space. The 3D model building and Fourier-snake procedure are demonstrated by segmenting deep structures of the human brain from MR volume data.",
    author = "G. Sz{\'e}kely and A. Kelemen and Ch Brechb{\"u}hler and G. Gerig",
    year = "1995",
    language = "English (US)",
    isbn = "9783540591207",
    volume = "905",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "495--505",
    booktitle = "Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings",

    }

    TY - GEN

    T1 - Segmentation of 3D gbjects from MRI volume data using constrained elastic deformations of flexible fourier surface models

    AU - Székely, G.

    AU - Kelemen, A.

    AU - Brechbühler, Ch

    AU - Gerig, G.

    PY - 1995

    Y1 - 1995

    N2 - This paper describes a new model-based segmentation technique combining desirable properties of physical models (snakes, [2]), shape representation by Fourier parametrization (Fourier snakes, [12]), and modelling of natural shape variability (eigenmodes, [7, 10]). Flexible shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation with rcspect to a small sct of stable landmarks (ACPC in our application) and explaining the remaining variability among a series of images with the model flexibility. Although straightforward, the extension to 3-D is severely impeded by finding a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization [16, 17] which achieves a uniform mapping between object surface and parameter space. The 3D model building and Fourier-snake procedure are demonstrated by segmenting deep structures of the human brain from MR volume data.

    AB - This paper describes a new model-based segmentation technique combining desirable properties of physical models (snakes, [2]), shape representation by Fourier parametrization (Fourier snakes, [12]), and modelling of natural shape variability (eigenmodes, [7, 10]). Flexible shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation with rcspect to a small sct of stable landmarks (ACPC in our application) and explaining the remaining variability among a series of images with the model flexibility. Although straightforward, the extension to 3-D is severely impeded by finding a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization [16, 17] which achieves a uniform mapping between object surface and parameter space. The 3D model building and Fourier-snake procedure are demonstrated by segmenting deep structures of the human brain from MR volume data.

    UR - http://www.scopus.com/inward/record.url?scp=84956852549&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84956852549&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:84956852549

    SN - 9783540591207

    VL - 905

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 495

    EP - 505

    BT - Computer Vision, Virtual Reality and Robotics in Medicine - 1st International Conference, CVRMed 1995, Proceedings

    PB - Springer Verlag

    ER -