Unencrypted login | home

Program Information

Proton Dose Calculations in Homogeneous Media

J Chapman

J Chapman1*, J Fontenot1,2, W Newhauser1,2, K Hogstrom1,2, (1) Louisiana State University, Baton Rouge, LA, (2) Mary Bird Perkins Cancer Center, Baton Rouge, LA

SU-E-T-535 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall

Purpose: To develop a pencil beam dose calculation algorithm for scanned proton beams that improves modeling of scatter events.

Methods: Our pencil beam algorithm (PBA) was developed for calculating dose from monoenergetic, parallel proton beams in homogeneous media. Fermi-Eyges theory was implemented for pencil beam transport. Elastic and nonelastic scatter effects were each modeled as a Gaussian distribution, with root mean square (RMS) widths determined from theoretical calculations and a nonlinear fit to a Monte Carlo (MC) simulated 1mm x 1mm proton beam, respectively. The PBA was commissioned using MC simulations in a flat water phantom. Resulting PBA calculations were compared with results of other models reported in the literature on the basis of differences between PBA and MC calculations of 80-20% penumbral widths. Our model was further tested by comparing PBA and MC results for oblique beams (45 degree incidence) and surface irregularities (step heights of 1 and 4 cm) for energies of 50-250 MeV and field sizes of 4cm x 4cm and 10cm x 10cm. Agreement between PBA and MC distributions was quantified by computing the percentage of points within 2% dose difference or 1mm distance to agreement.

Results: Our PBA improved agreement between calculated and simulated penumbral widths by an order of magnitude compared with previously reported values. For comparisons of oblique beams and surface irregularities, agreement between PBA and MC distributions was better than 99%.

Conclusions: Our algorithm showed improved accuracy over other models reported in the literature in predicting the overall shape of the lateral profile through the Bragg peak. This improvement was achieved by incorporating nonelastic scatter events into our PBA. The increased modeling accuracy of our PBA, incorporated into a treatment planning system, may improve the reliability of treatment planning calculations for patient treatments.

Contact Email