Addendum to the AAPM's TG-51 protocol for clinical reference dosimetry of high-energy photon beams

Abstract

An addendum to the AAPM's TG-51 protocol for the determination of absorbed dose to water in megavoltage photon beams is presented. This addendum continues the procedure laid out in TG-51 but new kQ data for photon beams, based on Monte Carlo simulations, are presented and recommendations are given to improve the accuracy and consistency of the protocol's implementation. The components of the uncertainty budget in determining absorbed dose to water at the reference point are introduced and the magnitude of each component discussed. Finally, the consistency of experimental determination of ND,w coefficients is discussed. It is expected that the implementation of this addendum will be straightforward, assuming that the user is already familiar with TG-51. The changes introduced by this report are generally minor, although new recommendations could result in procedural changes for individual users. It is expected that the effort on the medical physicist's part to implement this addendum will not be significant and could be done as part of the annual linac calibration.

FIG. 1.

Comparison of primary standards for ⁶⁰Co absorbed dose to water (as of the end of 2013). The uncertainty bars are given as the standard uncertainty and are the combined uncertainty for each standard compared to the BIPM standard. Data taken from BIPM's key comparison database (KCDB) database (www.bipm.org). This database is continually updated as comparisons are carried out.

FIG. 2.

k_Q factors for the NE2571 chamber as a function of beam-quality specifier %dd(10)_x. The references are those given in the report by Aalbers et al. (Ref. 102). Uncertainties are given as one standard uncertainty (Ref. 58). ⁶⁰Co is assigned a %dd(10)_x value of 58.4. A value of 85.0 represents the highest linac energy typically found in radiotherapy clinics. Figure courtesy of NCS.

FIG. 3.

Beam-quality data (Ref. 103) from six primary standard laboratories (individual laboratories are not identified) compared to the fit obtained by Kalach and Rogers (Ref. 107) for simulated data from heavily filtered clinical linac beams. The rms relative deviation from the fit is 0.7 %.