These tools will no longer be maintained as of December 31, 2024. Archived website can be found here. PubMed4Hh GitHub repository can be found here. Contact NLM Customer Service if you have questions.


Pubmed for Handhelds

PUBMED FOR HANDHELDS

Search MEDLINE/PubMed

  • Title: Imaging dose management using multi-resolution in CT-guided radiation therapy.
    Author: Sheng K, Jeraj R, Shaw R, Mackie TR, Paliwal BR.
    Journal: Phys Med Biol; 2005 Mar 21; 50(6):1205-19. PubMed ID: 15798317.
    Abstract:
    In image-guided radiation therapy, megavoltage computerized tomography (MVCT) delivers higher dose to the patient for lower image quality than diagnostic kilovoltage CT (kVCT). One way to reduce the mean imaging dose is to reduce the imaging volume, which is often sufficient for registration and dosimetry purposes. The filtered back projection using truncated data causes artefacts that degrade the image quality. Those artefacts can be effectively reduced by wavelet-based multi-resolution analysis (WMRA), in which the detail and approximate information are reconstructed separately to bypass the non-locality of filtered back projection. In this study, WMRA was used to reconstruct local images from both very low-dose kVCT scans from a bench-top tomotherapy unit and MVCT scans from helical tomotherapy. Results show that mean imaging dose can be significantly reduced by imaging a small region of interest. In simulation, the root-mean-square error brought by the truncation is smaller than 1-2% and depends on the level of dose reduction. On the other hand, the same mean dose that would have been delivered by a low-quality global CT can be conformed to a smaller volume to improve the visibility of low-contrast organs and fine structures using WMRA. Organs at risk can be avoided during repeated daily CT imaging when irregular-shaped reconstruction areas are used. WMRA does not involve computationally expensive iterations and is suitable for image-guided radiation therapy where imaging speed is essential. Compared with extrapolation methods, errors are further reduced to improve the detection of low contrast and fine structures.
    [Abstract] [Full Text] [Related] [New Search]