Publication: Automatic Differentiation for GPU-Accelerated 2D/3D Registration
Introduction
Applications
Tools
Research Groups
Workshops
Publications
   List Publications
   Advanced Search
   Info
   Add Publications
My Account
About

Automatic Differentiation for GPU-Accelerated 2D/3D Registration

- incollection -
 

Area
Biomedicine

Author(s)
Markus Grabner , Thomas Pock , Tobias Gross , Bernhard Kainz

Published in
Advances in Automatic Differentiation

Editor(s)
Christian H. Bischof, H. Martin Bücker, Paul D. Hovland, Uwe Naumann, J. Utke

Year
2008

Publisher
Springer

Abstract
A common task in medical image analysis is the alignment of data from different sources, e.g., X-ray images and computed tomography (CT) data. Such a task is generally known as registration. We demonstrate the applicability of automatic differentiation (ad) techniques to a class of 2D/3D registration problems which are highly computationally intensive and can therefore greatly benefit from a parallel implementation on recent graphics processing units (GPUs). However, being designed for graphics applications, GPUs have some restrictions which conflict with requirements for reverse mode ad, in particular for taping and TBR analysis. We discuss design and implementation issues in the presence of such restrictions on the target platform and present a method which can register a CT volume data set (512 512 288 voxels) with three X-ray images (512 512 pixels each) in 20 seconds on a GeForce 8800GTX graphics card.

Cross-References
Bischof2008AiA

BibTeX
@INCOLLECTION{
         Grabner2008ADf,
       author = "Markus Grabner and Thomas Pock and Tobias Gross and Bernhard Kainz",
       title = "Automatic Differentiation for {GPU}-Accelerated {2D/3D} Registration",
       doi = "10.1007/978-3-540-68942-3_23",
       pages = "259--269",
       abstract = "A common task in medical image analysis is the alignment of data from different
         sources, e.g., X-ray images and computed tomography (CT) data. Such a task is generally known as
         registration. We demonstrate the applicability of automatic differentiation (AD) techniques to a
         class of 2D/3D registration problems which are highly computationally intensive and can therefore
         greatly benefit from a parallel implementation on recent graphics processing units (GPUs). However,
         being designed for graphics applications, GPUs have some restrictions which conflict with
         requirements for reverse mode AD, in particular for taping and TBR analysis. We discuss design and
         implementation issues in the presence of such restrictions on the target platform and present a
         method which can register a CT volume data set (512 × 512 × 288 voxels) with three X-ray
         images (512 × 512 pixels each) in 20 seconds on a GeForce 8800GTX graphics card.",
       crossref = "Bischof2008AiA",
       booktitle = "Advances in Automatic Differentiation",
       publisher = "Springer",
       editor = "Christian H. Bischof and H. Martin B{\"u}cker and Paul D. Hovland and Uwe
         Naumann and J. Utke",
       isbn = "978-3-540-68935-5",
       issn = "1439-7358",
       year = "2008",
       ad_area = "Biomedicine"
}


back
  

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