Program Information
Impacts of Gantry Angle Dependent Scanning Beam Properties for Proton Treatment
Y Lin*, B Clasie , H Lu , J Flanz , K Jee , Massachusetts General Hospital and Harvard Medical School, Boston, MA
Presentations
SU-F-T-177 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose:In pencil beam scanning (PBS), the delivered spot MU, position and size are slightly different at different gantry angles. We investigated the level of delivery uncertainty at different gantry angles through a log file analysis.
Methods:34 PBS fields covering full 360 degrees gantry angle spread were collected retrospectively from 28 patients treated at our institution. All fields were delivered at zero gantry angle and the prescribed gantry angle, and measured at isocenter with the MatriXX 2D array detector at the prescribed gantry angle. The machine log files were analyzed to extract the delivered MU per spot and the beam position from the strip ionization chambers in the treatment nozzle. The beam size was separately measured as a function of gantry angle and beam energy. Using this information, the dose was calculated in a water phantom at both gantry angles and compared to the measurement using the 3D γ-index at 2mm/2%.
Results:The spot-by-spot difference between the beam position in the log files from the delivery at the two gantry angles has a mean of 0.3 and 0.4 mm and a standard deviation of 0.6 and 0.7 mm for x and y directions, respectively. Similarly, the spot-by-spot difference between the MU in the log files from the delivery at the two gantry angles has a mean 0.01% and a standard deviation of 0.7%. These small deviations lead to an excellent agreement in dose calculations with an average γ pass rate for all fields being approximately 99.7%. When each calculation is compared to the measurement, a high correlation in γ was also found.
Conclusion: Using machine logs files, we verified that PBS beam delivery at different gantry angles are sufficiently small and the planned spot position and MU. This study brings us one step closer to simplifying our patient-specific QA.
Contact Email: