With the recent interest in single fraction Stereotactic Body Radiation Therapy and the emerging prominence of the Rapid Arc radiotherapy technique capable of delivering a fast and accurate treatment, the in-field primary dose and out-of-field dose assessments are becoming increasingly important. Currently, full physics Monte Carlo calculations for dose calculations have been regarded as a ‘gold standard’ for dose assessments of the target and OAR (organ at risk). However, these Monte Carlo calculations require very long computation times. The current treatment planning methods provide shorter calculation times, but issues such as heterogeneities and model-based parameter calculations cause challenges and affect dose calculation accuracy. This thesis describes a new and fast dose estimation method leveraging parallel computing called EDK-SN, “Electron Dose Kernel-Discrete Ordinates”. This new method uses hybrid electron dose kernels driven by linear Boltzmann (discrete ordinates) photon transport method to carry out dose calculations. The method has proven effective for fast and accurate computations of out-of-field whole body dose calculations benchmarked to Monte Carlo with isotropic monoenergetic photon sources.
This thesis accomplishes adaptation of clinical Varian phase space data for use with general Monte Carlo codes including MCNP, and mapping accurate phase space data into the application of optimized EDK-SN dose calculation method with a 15-year-old patient phantom. The EDK-SN method with improved source term modeling is demonstrated to fall within accuracy of the measured golden beam data for a clinical water phantom.
| Identifer | oai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/54350 |
| Date | 07 January 2016 |
| Creators | Huang, Mi |
| Contributors | Wang, Chris K. |
| Publisher | Georgia Institute of Technology |
| Source Sets | Georgia Tech Electronic Thesis and Dissertation Archive |
| Language | en_US |
| Detected Language | English |
| Type | Dissertation |
| Format | application/pdf |
Page generated in 0.0016 seconds