Scanned carbon beam irradiation of moving films: comparison of measured and calculated response
1 GSI Helmholtzzentrum für Schwerionenforschung, Planckstr. 1, 64291 Darmstadt, Germany
2 TU Darmstadt, Institut für Festkörperphysik, Hochschulstr. 6, 64289 Darmstadt, Germany
3 Siemens AG, Healthcare Sector, Imaging & Therapy, Particle Therapy, Hofmannstr. 26, 91052 Erlangen, Germany
Radiation Oncology 2012, 7:55 doi:10.1186/1748-717X-7-55Published: 2 April 2012
Treatment of moving target volumes with scanned particle beams benefits from treatment planning that includes the time domain (4D). Part of 4D treatment planning is calculation of the expected result. These calculation codes should be verified against suitable measurements. We performed simulations and measurements to validate calculation of the film response in the presence of target motion.
All calculations were performed with GSI's treatment planning system TRiP. Interplay patterns between scanned particle beams and moving film detectors are very sensitive to slight deviations of the assumed motion parameters and therefore ideally suited to validate 4D calculations. In total, 14 film motion parameter combinations with lateral motion amplitudes of 8, 15, and 20 mm and 4 combinations for lateral motion including range changes were used. Experimental and calculated film responses were compared by relative difference, mean deviation in two regions-of-interest, as well as line profiles.
Irradiations of stationary films resulted in a mean relative difference of -1.52% ± 2.06% of measured and calculated responses. In comparison to this reference result, measurements with translational film motion resulted in a mean difference of -0.92% ± 1.30%. In case of irradiations incorporating range changes with a stack of 5 films as detector the deviations increased to -6.4 ± 2.6% (-10.3 ± 9.0% if film in distal fall-off is included) in comparison to -3.6% ± 2.5% (-13.5% ± 19.9% including the distal film) for the stationary irradiation. Furthermore, the comparison of line profiles of 4D calculations and experimental data showed only slight deviations at the borders of the irradiated area. The comparisons of pure lateral motion were used to determine the number of motion states that are required for 4D calculations depending on the motion amplitude. 6 motion states per 10 mm motion amplitude are sufficient to calculate the film response in the presence of motion.
By comparison to experimental data, the 4D extension of GSI's treatment planning system TRiP has been successfully validated for film response calculations in the presence of target motion within the accuracy limitation given by film-based dosimetry.