Spectral reconstruction of clinical megavoltage photon beams and the implications of spectral determination on the dosimetry of such beams.

ABSTRACT

An analysis technique, based on simulated annealing, is described which is employed to derive megavoltage photon beam spectral information from narrow beam attenuation measurements. Megavoltage photon beam spectra have been determined using this technique for linear accelerators from different manufacturers, and different models from individual manufacturers at a range of energies from nominal 6 MV to nominal 25 MV. All of the photon beams included in the study are in routine clinical use. The subsequent effects on dosimetry of employing derived primary spectra to specify beam quality are examined. The results suggest that the quality index TPR(20)10 may be insensitive to beam quality changes for high-energy beams in the range of 15 MV to 25 MV. Although the quality index may be insensitive as a beam quality specifier at these higher qualities, the actual difference in the calculated dose delivered using derived spectra as the quality specifier rather than TPR(20)10 is likely to be small, the results obtained indicating a difference of between 0.2% and 0.7% in the calculated dose delivered.