Noise properties of chord-image reconstruction

Dan Xia, Lifeng Yu, Emil Y. Sidky, Yu Zou, Nianming Zuo, Xiaochuan Pan

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

Recently, there has been much progress in algorithm development for image reconstruction in cone-beam computed tomography (CT). Current algorithms, including the chord-based algorithms, now accept minimal data sets for obtaining images on volume regions-of-interest (ROIs) thereby potentially allowing for reduction of X-ray dose in diagnostic CT. As these developments are relatively new, little effort has been directed at investigating the response of the resulting algorithm implementations to physical factors such as data noise. In this paper, we perform an investigation on the noise properties of ROI images reconstructed by using chord-based algorithms for different scanning configurations. We find that, for the cases under study, the chord-based algorithms yield images with comparable quality. Additionally, it is observed that, in many situations, large data sets contain extraneous data that may not reduce the ROI-image variances.

Original languageEnglish (US)
Pages (from-to)1328-1344
Number of pages17
JournalIEEE transactions on medical imaging
Volume26
Issue number10
DOIs
StatePublished - Oct 2007

Keywords

  • Chord
  • Computed tomography (CT)
  • Cone-beam CT
  • Noise
  • Reconstruction

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Noise properties of chord-image reconstruction'. Together they form a unique fingerprint.

Cite this