1 / 11

NAMIC Activities – Utah

NAMIC Activities – Utah. HD. AF. HNC. TBI. Image/Shape Analysis Volumetric tractography and DTI atlases Robust correspondences for shape Longitudinal shape analysis Segmentation Globally optimal surface estimation Fast, feature/shape-based image lookup Registration

hien
Télécharger la présentation

NAMIC Activities – Utah

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. NAMIC Activities – Utah HD AF HNC TBI • Image/Shape Analysis • Volumetric tractographyand DTI atlases • Robust correspondences for shape • Longitudinal shape analysis • Segmentation • Globally optimal surface estimation • Fast, feature/shape-based image lookup • Registration • Robust metrics for image match • Nonsmoothregularizers

  2. Volume Tractography • DTI Atlases to define ROIs • Goodletet al. • Volumetric tractography for white matter regions • Fletcher et al. • Group analysis on volumetric regions

  3. Robust Shape Correspondences • Datar et al. 2011 • Statistics of points and normals • Geodesic distances for point-to-point interactions

  4. Longitudinal Shape Parameterizationwith T. Fletcher, M. Datar Left atrium trend – before and after ablation • Mixed effects model • Hierarchical–properly accounts for staggered/missing data in individuals

  5. Segmentation • Graph-based image segmentation • Represent surfaces a min-cut in properly ordered graph • Liu et al. 2009 • Challenges • Objective functions that capture features, smoothness, coupled surfaces • Generalizations to 3D shapes

  6. Graph-Based Segmentation

  7. Fast Nearest Neighbor Lookup • Problem: from a large database of images, most similar images combine to form best segmentation • Label voting or nonparametric modeling paradigm • Especially important for heterogeneous data • Head&neck, cardiac • How to find similar shapes? • Deformation (slow) • Feature-based query (fast)

  8. Feature-Based Lookup • Strategy • Detect features and compare hierarchically • Pyramid matching (Grauman 2006)

  9. Robust Image Registration • Applications • Correspondence or coordinate system for comparing different individuals or times • E.g. longitudinal TBI, Afib before and after ablation • Segmentation from atlases or label voting • E.g. head and neck, endocardium from DCE • Challenges • Nonsmooth transformations • Singularities, tearing, sliding • Outliers/mismatches

  10. Registration Formulation • Strategy • Image match: apply robust versions of image metric (e.g. other norms) • Regularization: norms and operators that allow for nonsmooth (noninvertable) transformations • Issues • Well posedness • Optimization

  11. Registration Example

More Related