Dose reduction for kilovotage cone-beam computed tomography in radiation therapy

Jing Wang, Tianfang Li, Zhengrong Liang, Lei Xing

Research output: Contribution to journalArticle

78 Scopus citations

Abstract

Kilovotage cone-beam computed tomography (kV-CBCT) has shown potentials to improve the accuracy of a patient setup in radiotherapy. However, daily and repeated use of CBCT will deliver high extra radiation doses to patients. One way to reduce the patient dose is to lower mAs when acquiring projection data. This, however, degrades the quality of low mAs CBCT images dramatically due to excessive noises. In this work, we aim to improve the CBCT image quality from low mAs scans. Based on the measured noise properties of the sinogram, a penalized weighted least-squares (PWLS) objective function was constructed, and the ideal sinogram was then estimated by minimizing the PWLS objection function. To preserve edge information in the projection data, an anisotropic penalty term was designed using the intensity difference between neighboring pixels. The effectiveness of the presented algorithm was demonstrated by two experimental phantom studies. Noise in the reconstructed CBCT image acquired with a low mAs protocol was greatly suppressed after the proposed sinogram domain image processing, without noticeable sacrifice of the spatial resolution.

Original languageEnglish (US)
Pages (from-to)2897-2909
Number of pages13
JournalPhysics in medicine and biology
Volume53
Issue number11
DOIs
StatePublished - Jun 7 2008

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Dose reduction for kilovotage cone-beam computed tomography in radiation therapy'. Together they form a unique fingerprint.

  • Cite this