Encrypted login | home

Program Information

Raystation Electron Monte Carlo Commissioning and Clinical Implementation

no image available
C Allen

C Allen*, P Sansourekidou , D Pavord , Health-quest, Poughkeepsie, NY

Presentations

SU-E-T-277 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose:
To evaluate the Raystation v4.0 Electron Monte Carlo algorithm for an Elekta Infinity linear accelerator and commission for clinical use.
Methods:
A total of 199 tests were performed (75 Export and Documentation, 20 PDD, 30 Profiles, 4 Obliquity, 10 Inhomogeneity, 55 MU Accuracy, and 5 Grid and Particle History). Export and documentation tests were performed with respect to MOSAIQ (Elekta AB) and RadCalc (Lifeline Software Inc). Mechanical jaw parameters and cutout magnifications were verified. PDD and profiles for open cones and cutouts were extracted and compared with water tank measurements. Obliquity and inhomogeneity for bone and air calculations were compared to film dosimetry. MU calculations for open cones and cutouts were performed and compared to both RadCalc and simple hand calculations. Grid size and particle histories were evaluated per energy for statistical uncertainty performance. Acceptability was categorized as follows: performs as expected, negligible impact on workflow, marginal impact, critical impact or safety concern, and catastrophic impact of safety concern.
Results:
Overall results are: 88.8% perform as expected, 10.2% negligible, 2.0% marginal, 0% critical and 0% catastrophic. Results per test category are as follows: Export and Documentation: 100% perform as expected, PDD: 100% perform as expected, Profiles: 66.7% perform as expected, 33.3% negligible, Obliquity: 100% marginal, Inhomogeneity 50% perform as expected, 50% negligible, MU Accuracy: 100% perform as expected, Grid and particle histories: 100% negligible. To achieve distributions with satisfactory smoothness level, 5,000,000 particle histories were used. Calculation time was approximately 1 hour.
Conclusion:
Raystation electron Monte Carlo is acceptable for clinical use. All of the issues encountered have acceptable workarounds. Known issues were reported to Raysearch and will be resolved in upcoming releases.



Contact Email: