An Optimum Normalization Technique for Electron Monte Carlo Treatment Planning
Gregory K. Bartlett1*, Xiaoyi Lu2, John Kent3, Colleen DesRosiers4, Yuichi Akino5, Indra Das6, (1) Indiana Univeristy Health University Hospital, Indianapolis, Indiana ,,,(2) Indiana University School of Medicine, Indianapolis, Indiana, (3) Indiana University Health Methodist Hospital, Indianapolis, IN, (4) University Hospital, Indianapolis , IN, (5) Osaka University Graduate School of Medicine, Osaka, Japan ,,,(6) Indiana University- School of Medicine, Indianapolis, INSU-E-T-541 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To evaluate the optimum normalization technique using the treatment planning system (TPS) to calculate dose with Electron Monte Carlo (eMC) algorithms.
Methods: Eclipse TPS was used to generate 6 and 9 MeV plans for field apertures ranging from 3 cm to 10 cm, calculation grid sizes from 1 mm to 5mm, source to surface distance (SSD) from 100 to 110, and beam incident angles from 0 to 30 degrees. Each plan was normalized to the (1) TPS global maximum and (2) the physical dmax depth on the central axis given by measured percent depth dose (PDD) data. To validate the eMC relative dose distributions, PDD and beam profiles (at dmax) were obtained in a water phantom. An irregular surface phantom and a small domed phantom were constructed for dose measurements.
Results: In all cases, the TPS' PDD agreed with the measured PDD within 2.5 % beyond a depth of 0.5cm. Absolute dose measurements on the irregular surface and domed phantoms also agreed with an average difference of 2.5%. Normalizing to the physical dmax on the central axis resulted in MUs up to 10% higher than normalizing to TPS global maximum. This was due to the eMC dmax shifting as much as 6mm shallower in depth, or shifting away from the central axis vary based on field apertures and beam angle incidences. MU calculations based on tabular data from beam commissioning differed from measured by 11.1% on average when oblique angles and cutouts were used. The tabulated data are based on measurements performed in a flat phantom with no beam obliquity.
Conclusion: The planned MUs using the global maximum are more closely matched with the measured values as compared with MUs generated from tabulated data. Thus, the TPS generated MUs should more accurately reflect dose delivered for patient treatments.