Publication: Computational Strategies for Meshfree Nonrigid Registration
Introduction
Applications
Tools
Research Groups
Workshops
Publications
   List Publications
   Advanced Search
   Info
   Add Publications
My Account
About

Computational Strategies for Meshfree Nonrigid Registration

- Ph.D. thesis -
 

Area
Biomedicine, Optimization

Author(s)
Eliezer Kahn

Year
2006

Abstract
Biological shapes such as the brain are difficult to register due to their complicated geometry. To deal with this, registration methods often rely on a transformation model consisting of a dense regular grid such as a free form deformation or B-spline grid. However, very dense grids or meshes are usually needed to register images with convoluted shapes, and a regular mesh structure is not well suited for the irregular structure of the brain. What is therefore needed is a meshfree approach such as a radial basis function transformation model. Unfortunately, because radial basis functions are typically non-compact, using them with large numbers of points is fraught with numerical difficulties and, as a result, their use in image registration is not prevalent. The goal of this work is to overcome these computational difficulties so that radial basis function transformations can be used efficiently, even with large numbers of points. To achieve this, a new registration framework was developed based on automatic differentiation and the fast multipole method. Automatic differentiation is useful since an important component of registration is computing the gradient of the similarity metric which is to be optimized. Automatic differentiation allows one to efficiently calculate gradients without having to write any gradient code explicitly. Although the technique of automatic differentiation is well established, it does not appear to be used for image registration. The fast multipole method was developed to efficiently evaluate large sums such as radial basis functions but its use in image registration is still minimal. With the integration of these algorithms within a complete registration framework, it should be possible to obtain a truly meshfree registration.

AD Tools
ADOL-C, Treeverse / Revolve

AD Theory and Techniques
Checkpointing, Reverse Mode

BibTeX
@PHDTHESIS{
         Kahn2006CSf,
       author = "Eliezer Kahn",
       title = "Computational Strategies for Meshfree Nonrigid Registration",
       school = "Yale University",
       month = "December",
       year = "2006",
       ad_tools = "ADOL-C, Treeverse / Revolve",
       ad_area = "Biomedicine, Optimization",
       ad_theotech = "Checkpointing, Reverse Mode",
       url = "http://noodle.med.yale.edu/~kahn/",
       abstract = "Biological shapes such as the brain are difficult to register due to their
         complicated geometry. To deal with this, registration methods often rely on a transformation model
         consisting of a dense regular grid such as a free form deformation or B-spline grid. However, very
         dense grids or meshes are usually needed to register images with convoluted shapes, and a regular
         mesh structure is not well suited for the irregular structure of the brain. What is therefore needed
         is a meshfree approach such as a radial basis function transformation model. Unfortunately, because
         radial basis functions are typically non-compact, using them with large numbers of points is fraught
         with numerical difficulties and, as a result, their use in image registration is not prevalent. The
         goal of this work is to overcome these computational difficulties so that radial basis function
         transformations can be used efficiently, even with large numbers of points. To achieve this, a new
         registration framework was developed based on automatic differentiation and the fast multipole
         method. Automatic differentiation is useful since an important component of registration is
         computing the gradient of the similarity metric which is to be optimized. Automatic differentiation
         allows one to efficiently calculate gradients without having to write any gradient code explicitly.
         Although the technique of automatic differentiation is well established, it does not appear to be
         used for image registration. The fast multipole method was developed to efficiently evaluate large
         sums such as radial basis functions but its use in image registration is still minimal. With the
         integration of these algorithms within a complete registration framework, it should be possible to
         obtain a truly meshfree registration."
}


back
  

Contact:
autodiff.org
Username:
Password:
(lost password)