A comparison of beam characteristics for gated and nongated clinical x-ray beams.

Respiratory gating has only recently been applied to conventional external beam radiotherapy. In order for respiratory gating to be used clinically, an evaluation of the dosimetric effects of small units of delivered dose must be performed. The purpose of this study is to systematically evaluate the effect of various gating sequences on x-ray central axis output, ionization ratios (nominal accelerating potential), beam flatness, and beam symmetry. Measurements were taken for 6 and 18 MV photons on a linear accelerator that generates the gate by using a gridded electron gun to stop the electron flow to the wave-guide. The beam output, energy, flatness, and symmetry did not vary by more than 0.8 percent in most of the gating sequences. The maximum output deviations (0.8 percent), flatness deviations (1.9 percent), and symmetry deviations (0.8 percent) occurred when a low number of monitor units (<5 MU) were delivered in the gating window. Although these deviations are not clinically significant, each linear accelerator should be evaluated carefully before clinical implementation.

Dosimetric verification of two commercially available three-dimensional treatment planning systems using the TG 23 test package.

The Task Group 23 (TG-23) radiation treatment planning dosimetry verification package was used to evaluate the dosimetric accuracy of two commercially available treatment planning systems. The TG-23 test package contains experimentally measured beam data for two x-ray beams (4 and 18 MV) that can be used as input for 3D-RTP (three-dimensional radiation treatment planning) systems. Once the beam data is entered and modeled, a series of test cases are performed that isolate different aspects of the dose computational process. The computed values from the 3D-RTP system are compared against the measured dosimetry data, included in the package, for a set of comparison points within each test case. Both of the treatment planning systems that were studied provided excellent agreement between computed and measured doses. The cumulative 4 and 18 MV TG-23 test results for the convolution/superposition based planning system indicates that 96% of the dosimetric test points are within +/-2%, and 98% are within +/-3% of the tabulated TG-23 values. The dosimetric TG-23 test results for the pencil beam kernel based planning system are similar, with 96% of the test points falling within +/-2%, and 99% falling within +/-3% of the TG-23 measurements.