Investigation of practical approaches to evaluating cumulative dose for cone beam computed tomography (CBCT) from standard CT dosimetry measurements: a Monte Carlo study

A function called G(x)(L) was introduced by the International Commission on Radiation Units and Measurements (ICRU) Report-87 to facilitate measurement of cumulative dose for CT scans within long phantoms as recommended by the American Association of Physicists in Medicine (AAPM) TG-111. The G(x)(L) function is equal to the ratio of the cumulative dose at the middle of a CT scan to the volume weighted CTDI (CTDIvol), and was investigated for conventional multi-slice CT scanners operating with a moving table. As the stationary table mode, which is the basis for cone beam CT (CBCT) scans, differs from that used for conventional CT scans, the aim of this study was to investigate the extension of the G(x)(L) function to CBCT scans. An On-Board Imager (OBI) system integrated with a TrueBeam linac was simulated with Monte Carlo EGSnrc/BEAMnrc, and the absorbed dose was calculated within PMMA, polyethylene (PE), and water head and body phantoms using EGSnrc/DOSXYZnrc, where the body PE body phantom emulated the ICRU/AAPM phantom. Beams of width 40-500 mm and beam qualities at tube potentials of 80-140 kV were studied. Application of a modified function of beam width (W) termed G(x)(W), for which the cumulative dose for CBCT scans f(0) is normalized to the weighted CTDI (CTDIw) for a reference beam of width 40 mm, was investigated as a possible option. However, differences were found in G(x)(W) with tube potential, especially for body phantoms, and these were considered to be due to differences in geometry between wide beams used for CBCT scans and those for conventional CT. Therefore, a modified function G(x)(W)(100) has been proposed, taking the form of values of f(0) at each position in a long phantom, normalized with respect to dose indices f(100)(150)(x) measured with a 100 mm pencil ionization chamber within standard 150 mm PMMA phantoms, using the same scanning parameters, beam widths and positions within the phantom. f(100)(150)(x) averages the dose resulting from a CBCT scan over the 100 mm length. Like the G(x)(L) function, the G(x)(W)(100) function showed only a weak dependency on tube potential at most positions for the phantoms studied. The results were fitted to polynomial equations from which f (0) within the longer PMMA, PE, or water phantoms can be evaluated from measurements of f(100)(150)(x). Comparisons with other studies, suggest that these functions may be suitable for application to any CT or CBCT scan acquired with stationary table mode.