An X-ray Source Model and Characterization Method for Computing kV Radiation Dose

Date
2014-05-26
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
According to the American Association of Physicists in Medicine Task Group (AAPM) 75, it is the responsibility of the medical physics community to assess, reduce, and optimize the imaging dose delivered during image-guided radiation therapy. It is currently difficult to accurately estimate patient-specific spatial dose distributions deposited by kilovoltage (kV) cone-beam CT (CBCT) using existing techniques. This thesis describes the development and validation of a characterization method and virtual point source model for describing kV radiographic and CBCT x-ray imaging units for our in-house kV dose computation software, kVDoseCalc. kV CBCT beams have spatial and spectral fluence variations. In this thesis, the spectrum is characterized using the accelerating potential (kVp) and half-value layer (HVL). These are matched to spectra generated by third-party freeware. The fluence is then isolated from measured in-air kerma. We assume that the fluence is a function of three separate independent variables: the transverse-axis spectra; and the transverse- and radial-axis fluence profiles. This work demonstrates that kV x-ray sources can be characterized using in-air ionization chamber measurements of HVL and kerma profiles. The characterization method and model are validated by comparing the absorbed dose computed by kVDoseCalc with measurements taken using ionization chambers and thermoluminescent detectors. Doses were measured in homogeneous and heterogeneous block phantoms for radiographic imaging procedures. We then measured imaging doses in homogeneous cylindrical and heterogeneous anthropomorphic phantoms for default CBCT protocols. The percent dose difference between measurement and computation was generally ≤3% in homogeneous and ≤6% in heterogeneous geometries, respectively. In the anthropomorphic phantom, the average percent difference ranged from 3–6% depending on the imaging protocol and site. The described method makes the accurate computation of radiographic and CBCT absorbed dose possible. This method has since been used by others in our group to accurately characterize a conventional CT scanner, superficial/orthovoltage therapy units, and to characterize an in-vivo film dosimetry program in our centre. Our technique is notable for relying only on empirical measurements. It allows a more accurate kV dose computation than treatment planning systems adapted for this purpose and is more clinically feasible than conventional Monte Carlo simulations.
Description
Keywords
Physics--Radiation
Citation
Poirier, Y. (2014). An X-ray Source Model and Characterization Method for Computing kV Radiation Dose (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/24942