O13. A comparison of TPS commissioning dosimeters

Abstract

Introduction The quality of treatment planning system (TPS) commissioning data largely determines the quality of the derived TPS beam model. Modern treatment techniques require a very high degree of competence from the dosimetrist, as well as the dosimetry equipment used for TPS commissioning. Once derived, the beam model should be verified with a range of tests other than commissioning procedures to test the beam model against linac output. Generally, dosimeters used for TPS beam data collection differ largely from those used for post-modelling verification measurements; therefore their correlation needs to be investigated. Materials and Methods Reference beam profile commissioning data were acquired with a Diamond detector (PTW-60019) for field sizes 2 × 2, 3 × 3 and 5 × 5 cm2 at depths 5 and 10 cm in water. Measurements were repeated with a 2D diode array (SunNuclear Mapcheck2) and compared to reference data. Default array spatial resolution of 1 cm was optimized to 1 mm by manual stepping of the device; therefore only measured (non-interpolated) data was gathered and collated using Excel VBA. A 2D gamma analysis with criteria 1%/1 mm was used for quantitative comparisons. Results Comparison of datasets revealed γ 1 in the open and penumbra regions for all scans evaluated. Mapcheck generally yielded a sharper penumbra (0.5 mm) while the largest differences were seen outside the field with Mapcheck demonstrating a 1.5% larger relative dose. Such small deviations can be attributed to differences in the detector sensitive volumes and dosimeter setup. Further, reducing the amount of steps per array profile by two-thirds (from 10 step increments to 3) did not influence the results significantly. Conclusion Results showed with increased resolution the sensitivity of the array are comparable to a diamond detector for small field profiles. Excellent overall agreement was found between these. However, the dosimetric accomplishment may be weakened by acquisition and processing workload therefore optimization thereof was demonstrated.