Statistics of pose and shape in multi-object complexes using principal geodesic analysis

Martin Styner, Kevin Gorczowski, Tom Fletcher, Ja Yeon Jeong, Stephen M. Pizer, Guido Gerig

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

    Abstract

    A main focus of statistical shape analysis is the description of variability of a population of geometric objects. In this paper, we present work in progress towards modeling the shape and pose variability of sets of multiple objects. Principal geodesic analysis (PGA) is the extension of the standard technique of principal component analysis (PCA) into the nonlinear Riemannian symmetric space of pose and our medial m-rep shape description, a space in which use of PCA would be incorrect. In this paper, we discuss the decoupling of pose and shape in multi-object sets using different normalization settings. Further, we introduce new methods of describing the statistics of object pose using a novel extension of PGA, which previously has been used for global shape statistics. These new pose statistics are then combined with shape statistics to form a more complete description of multi-object complexes. We demonstrate our methods in an application to a longitudinal pediatric autism study with object sets of 10 subcortical structures in a population of 20 subjects. The results show that global scale accounts for most of the major mode of variation across time. Furthermore, the PGA components and the corresponding distribution of different subject groups vary significantly depending on the choice of normalization, which illustrates the importance of global and local pose alignment in multi-object shape analysis.

    Original languageEnglish (US)
    Title of host publicationMedical Imaging and Augmented Reality - Third International Workshop
    Pages1-8
    Number of pages8
    Volume4091 LNCS
    StatePublished - 2006
    Event3rd International Workshop on Medical Imaging and Augmented Reality - Shanghai, China
    Duration: Aug 17 2006Aug 18 2006

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume4091 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other3rd International Workshop on Medical Imaging and Augmented Reality
    CountryChina
    CityShanghai
    Period8/17/068/18/06

    Fingerprint

    Geodesic
    Principal Component Analysis
    Statistics
    Principal component analysis
    Shape Analysis
    Normalization
    Pediatrics
    Autistic Disorder
    Riemannian Symmetric Space
    Geometric object
    Population
    Decoupling
    Statistical Analysis
    Alignment
    Vary
    Modeling
    Demonstrate
    Object

    ASJC Scopus subject areas

    • Computer Science(all)
    • Biochemistry, Genetics and Molecular Biology(all)
    • Theoretical Computer Science

    Cite this

    Styner, M., Gorczowski, K., Fletcher, T., Jeong, J. Y., Pizer, S. M., & Gerig, G. (2006). Statistics of pose and shape in multi-object complexes using principal geodesic analysis. In Medical Imaging and Augmented Reality - Third International Workshop (Vol. 4091 LNCS, pp. 1-8). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4091 LNCS).

    Statistics of pose and shape in multi-object complexes using principal geodesic analysis. / Styner, Martin; Gorczowski, Kevin; Fletcher, Tom; Jeong, Ja Yeon; Pizer, Stephen M.; Gerig, Guido.

    Medical Imaging and Augmented Reality - Third International Workshop. Vol. 4091 LNCS 2006. p. 1-8 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4091 LNCS).

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

    Styner, M, Gorczowski, K, Fletcher, T, Jeong, JY, Pizer, SM & Gerig, G 2006, Statistics of pose and shape in multi-object complexes using principal geodesic analysis. in Medical Imaging and Augmented Reality - Third International Workshop. vol. 4091 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4091 LNCS, pp. 1-8, 3rd International Workshop on Medical Imaging and Augmented Reality, Shanghai, China, 8/17/06.
    Styner M, Gorczowski K, Fletcher T, Jeong JY, Pizer SM, Gerig G. Statistics of pose and shape in multi-object complexes using principal geodesic analysis. In Medical Imaging and Augmented Reality - Third International Workshop. Vol. 4091 LNCS. 2006. p. 1-8. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    Styner, Martin ; Gorczowski, Kevin ; Fletcher, Tom ; Jeong, Ja Yeon ; Pizer, Stephen M. ; Gerig, Guido. / Statistics of pose and shape in multi-object complexes using principal geodesic analysis. Medical Imaging and Augmented Reality - Third International Workshop. Vol. 4091 LNCS 2006. pp. 1-8 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{26c58823f358492aa024ec32780bb81e,
    title = "Statistics of pose and shape in multi-object complexes using principal geodesic analysis",
    abstract = "A main focus of statistical shape analysis is the description of variability of a population of geometric objects. In this paper, we present work in progress towards modeling the shape and pose variability of sets of multiple objects. Principal geodesic analysis (PGA) is the extension of the standard technique of principal component analysis (PCA) into the nonlinear Riemannian symmetric space of pose and our medial m-rep shape description, a space in which use of PCA would be incorrect. In this paper, we discuss the decoupling of pose and shape in multi-object sets using different normalization settings. Further, we introduce new methods of describing the statistics of object pose using a novel extension of PGA, which previously has been used for global shape statistics. These new pose statistics are then combined with shape statistics to form a more complete description of multi-object complexes. We demonstrate our methods in an application to a longitudinal pediatric autism study with object sets of 10 subcortical structures in a population of 20 subjects. The results show that global scale accounts for most of the major mode of variation across time. Furthermore, the PGA components and the corresponding distribution of different subject groups vary significantly depending on the choice of normalization, which illustrates the importance of global and local pose alignment in multi-object shape analysis.",
    author = "Martin Styner and Kevin Gorczowski and Tom Fletcher and Jeong, {Ja Yeon} and Pizer, {Stephen M.} and Guido Gerig",
    year = "2006",
    language = "English (US)",
    isbn = "3540372202",
    volume = "4091 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "1--8",
    booktitle = "Medical Imaging and Augmented Reality - Third International Workshop",

    }

    TY - GEN

    T1 - Statistics of pose and shape in multi-object complexes using principal geodesic analysis

    AU - Styner, Martin

    AU - Gorczowski, Kevin

    AU - Fletcher, Tom

    AU - Jeong, Ja Yeon

    AU - Pizer, Stephen M.

    AU - Gerig, Guido

    PY - 2006

    Y1 - 2006

    N2 - A main focus of statistical shape analysis is the description of variability of a population of geometric objects. In this paper, we present work in progress towards modeling the shape and pose variability of sets of multiple objects. Principal geodesic analysis (PGA) is the extension of the standard technique of principal component analysis (PCA) into the nonlinear Riemannian symmetric space of pose and our medial m-rep shape description, a space in which use of PCA would be incorrect. In this paper, we discuss the decoupling of pose and shape in multi-object sets using different normalization settings. Further, we introduce new methods of describing the statistics of object pose using a novel extension of PGA, which previously has been used for global shape statistics. These new pose statistics are then combined with shape statistics to form a more complete description of multi-object complexes. We demonstrate our methods in an application to a longitudinal pediatric autism study with object sets of 10 subcortical structures in a population of 20 subjects. The results show that global scale accounts for most of the major mode of variation across time. Furthermore, the PGA components and the corresponding distribution of different subject groups vary significantly depending on the choice of normalization, which illustrates the importance of global and local pose alignment in multi-object shape analysis.

    AB - A main focus of statistical shape analysis is the description of variability of a population of geometric objects. In this paper, we present work in progress towards modeling the shape and pose variability of sets of multiple objects. Principal geodesic analysis (PGA) is the extension of the standard technique of principal component analysis (PCA) into the nonlinear Riemannian symmetric space of pose and our medial m-rep shape description, a space in which use of PCA would be incorrect. In this paper, we discuss the decoupling of pose and shape in multi-object sets using different normalization settings. Further, we introduce new methods of describing the statistics of object pose using a novel extension of PGA, which previously has been used for global shape statistics. These new pose statistics are then combined with shape statistics to form a more complete description of multi-object complexes. We demonstrate our methods in an application to a longitudinal pediatric autism study with object sets of 10 subcortical structures in a population of 20 subjects. The results show that global scale accounts for most of the major mode of variation across time. Furthermore, the PGA components and the corresponding distribution of different subject groups vary significantly depending on the choice of normalization, which illustrates the importance of global and local pose alignment in multi-object shape analysis.

    UR - http://www.scopus.com/inward/record.url?scp=33749381652&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33749381652&partnerID=8YFLogxK

    M3 - Conference contribution

    SN - 3540372202

    SN - 9783540372202

    VL - 4091 LNCS

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

    SP - 1

    EP - 8

    BT - Medical Imaging and Augmented Reality - Third International Workshop

    ER -