Translator Disclaimer
Paper
13 March 2009 Prior-image-based few-view cone beam CT for applications to daily scan in image-guided radiation therapy: preliminary study
Author Affiliations +
Proceedings Volume 7258, Medical Imaging 2009: Physics of Medical Imaging; 72581U (2009) https://doi.org/10.1117/12.811345
Event: SPIE Medical Imaging, 2009, Lake Buena Vista (Orlando Area), Florida, United States
Abstract
Interfraction motion of a treatment target such as the prostate in radiation therapy (RT) is, in part, responsible for large planning target volume (PTV) margins and related side effects. Online adjustment of the treatment based on timely cone-beam CT (CBCT) images can be particularly useful for patients with large interfraction motion. However, radiation dose to the patient due to frequent CBCT poses a radiation safety concern. One unique feature of CBCT for interfraction motion detection is the availability of a prior anatomical image most of which has not changed. We propose an iterative algorithm, for image reconstruction from a very limited number of projections in CBCT, that is based on total variation (TV) minimization subject to the constraints of data fidelity and positivity and that utilizes anatomical image prior information. Numerical studies for a 2D fan-beam geometry suggests the proposed algorithm can potentially contribute to lowering the radiation dose to the patient by allowing satisfactory image reconstruction from a very limited number of projections.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Seungryong Cho, Erik Pearson, Emil Y. Sidky, Junguo Bian, Charles A. Pelizzari, and Xiaochuan Pan "Prior-image-based few-view cone beam CT for applications to daily scan in image-guided radiation therapy: preliminary study", Proc. SPIE 7258, Medical Imaging 2009: Physics of Medical Imaging, 72581U (13 March 2009); https://doi.org/10.1117/12.811345
PROCEEDINGS
7 PAGES


SHARE
Advertisement
Advertisement
Back to Top