A self-adapting fast reconstruction method for high resolution images in cone beam CT

Mingjun Li, Dinghua Zhang, Kuidong Huang, Qingchao Yu, Shunli Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Reconstruction speed and image quality are the keys of 3D image reconstruction in cone beam CT (CBCT), this article have researched on the 3D image reconstruction in CBCT, and a self-adapting fast reconstruction method for high resolution images is presented. The method is based on the FDK algorithm and Z-line data first algorithm, which introduces the reconstruction task division algorithm based on Z-line truncation, and load algorithm of filtered projection images based on the VOI reconstruction region. It divide the reconstruction task in a self-adapting method according to the task amount and memory size, thus reduce the data load amount and frequency of filtered projection images. The experimental results show that the method achieves fast reconstruction speed for high resolution images in CBCT, and the image quality almost has no difference with FDK algorithm.

Original languageEnglish
Title of host publicationProceedings of the 5th International Conference on Image and Graphics, ICIG 2009
PublisherIEEE Computer Society
Pages23-27
Number of pages5
ISBN (Print)9780769538839
DOIs
StatePublished - 2009
Event5th International Conference on Image and Graphics, ICIG 2009 - Xi'an, Shanxi, China
Duration: 20 Sep 200923 Sep 2009

Publication series

NameProceedings of the 5th International Conference on Image and Graphics, ICIG 2009

Conference

Conference5th International Conference on Image and Graphics, ICIG 2009
Country/TerritoryChina
CityXi'an, Shanxi
Period20/09/0923/09/09

Keywords

  • Cone beam CT
  • FDK
  • High resolution
  • Reconstruction
  • Self-adapting
  • Z-line data first algorithm

Fingerprint

Dive into the research topics of 'A self-adapting fast reconstruction method for high resolution images in cone beam CT'. Together they form a unique fingerprint.

Cite this