Premium
Sci—Thur AM: YIS ‐ 09: Validation of a General Empirically‐Based Beam Model for kV X‐ray Sources
Author(s) -
Poirier Y.,
Sommerville M.,
Johnstone C.D.,
Gräfe J.,
Nygren I.,
Khan R.,
VillarealBarajas J.E.,
Jacso F.,
Tambasco M.
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4894971
Subject(s) - imaging phantom , thermoluminescent dosimeter , nuclear medicine , cone beam computed tomography , radiography , physics , beam (structure) , fluence , ionization chamber , x ray , optics , dosimetry , computed tomography , ionization , medicine , radiology , nuclear physics , ion , laser , quantum mechanics , dosimeter
Purpose: To present an empirically‐based beam model for computing dose deposited by kilovoltage (kV) x‐rays and validate it for radiographic, CT, CBCT, superficial, and orthovoltage kV sources. Method and Materials: We modeled a wide variety of imaging (radiographic, CT, CBCT) and therapeutic (superficial, orthovoltage) kV x‐ray sources. The model characterizes spatial variations of the fluence and spectrum independently. The spectrum is derived by matching measured values of the half value layer (HVL) and nominal peak potential (kVp) to computationally‐derived spectra while the fluence is derived from in‐air relative dose measurements. This model relies only on empirical values and requires no knowledge of proprietary source specifications or other theoretical aspects of the kV x‐ray source. To validate the model, we compared measured doses to values computed using our previously validated in‐house kV dose computation software, kVDoseCalc. The dose was measured in homogeneous and anthropomorphic phantoms using ionization chambers and LiF thermoluminescent detectors (TLDs), respectively. Results: The maximum difference between measured and computed dose measurements was within 2.6%, 3.6%, 2.0%, 4.8%, and 4.0% for the modeled radiographic, CT, CBCT, superficial, and the orthovoltage sources, respectively. In the anthropomorphic phantom, the computed CBCT dose generally agreed with TLD measurements, with an average difference and standard deviation ranging from 2.4 ± 6.0% to 5.7 ± 10.3% depending on the imaging technique. Most (42/62) measured TLD doses were within 10% of computed values. Conclusions: The proposed model can be used to accurately characterize a wide variety of kV x‐ray sources using only empirical values.