Drexel University Home Pagewww.drexel.edu DREXEL UNIVERSITY LIBRARIES HOMEPAGE >>
iDEA DREXEL ARCHIVES >>

iDEA: Drexel E-repository and Archives > Drexel Theses and Dissertations > Drexel Theses and Dissertations > GPU-accelerated deformable registration of cone-beam CT images: design and implementation in the compute unified device architecture

Please use this identifier to cite or link to this item: http://hdl.handle.net/1860/2865

Title: GPU-accelerated deformable registration of cone-beam CT images: design and implementation in the compute unified device architecture
Authors: Jain, Nakul
Keywords: Electric engineering;Tomography;Computer architecture
Issue Date: 6-Aug-2008
Abstract: This thesis describes a stream-processing model for the demons deformable registration algorithm, suitable for use on a graphics processing unit (GPU). The streaming version of the demons algorithm is developed using the Compute Unified Device Architecture (CUDA) and executed on NVidia‘s 8800 GTS GPU. Performance results using cone-beam CT data of a preserved swine lung indicate that the GPU-based implementation achieves a substantial speedup of 60 times when compared to an optimized reference implementation on a 2.8 GHz Intel processor. In addition, the accuracy of the GPU-based implementation was found to be excellent, with an RMS difference of less than 0.1 mm when compared with the reference implementation.
URI: http://hdl.handle.net/1860/2865
Appears in Collections:Drexel Theses and Dissertations

Files in This Item:

File Description SizeFormat
Jain_Nakul.pdf829.61 kBAdobe PDFView/Open
View Statistics

Items in iDEA are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! iDEA Software Copyright © 2002-2010  Duraspace - Feedback