Geodesic image regression with a sparse parameterization of diffeomorphisms

James Fishbaugh, Marcel Prastawa, Guido Gerig, Stanley Durrleman

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

    Abstract

    Image regression allows for time-discrete imaging data to be modeled continuously, and is a crucial tool for conducting statistical analysis on longitudinal images. Geodesic models are particularly well suited for statistical analysis, as image evolution is fully characterized by a baseline image and initial momenta. However, existing geodesic image regression models are parameterized by a large number of initial momenta, equal to the number of image voxels. In this paper, we present a sparse geodesic image regression framework which greatly reduces the number of model parameters. We combine a control point formulation of deformations with a L1 penalty to select the most relevant subset of momenta. This way, the number of model parameters reflects the complexity of anatomical changes in time rather than the sampling of the image. We apply our method to both synthetic and real data and show that we can decrease the number of model parameters (from the number of voxels down to hundreds) with only minimal decrease in model accuracy. The reduction in model parameters has the potential to improve the power of ensuing statistical analysis, which faces the challenging problem of high dimensionality.

    Original languageEnglish (US)
    Title of host publicationGeometric Science of Information - First International Conference, GSI 2013, Proceedings
    Pages95-102
    Number of pages8
    DOIs
    StatePublished - 2013
    Event1st International SEE Conference on Geometric Science of Information, GSI 2013 - Paris, France
    Duration: Aug 28 2013Aug 30 2013

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume8085 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Other

    Other1st International SEE Conference on Geometric Science of Information, GSI 2013
    CountryFrance
    CityParis
    Period8/28/138/30/13

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Fingerprint Dive into the research topics of 'Geodesic image regression with a sparse parameterization of diffeomorphisms'. Together they form a unique fingerprint.

  • Cite this

    Fishbaugh, J., Prastawa, M., Gerig, G., & Durrleman, S. (2013). Geodesic image regression with a sparse parameterization of diffeomorphisms. In Geometric Science of Information - First International Conference, GSI 2013, Proceedings (pp. 95-102). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8085 LNCS). https://doi.org/10.1007/978-3-642-40020-9_9