Thanaphat Chongsan.. Dosimetric investigation of Proton therapy on CT-based patient data using Monte Carlo Simulation. Master's Degree(Medical Physics). Mahidol University. Mahidol University Library and Knowledge Center. : Mahidol University, 2016.
Dosimetric investigation of Proton therapy on CT-based patient data using Monte Carlo Simulation
Abstract:
The aim of radiotherapy is to deliver high radiation dose to the tumor with low radiation dose to healthy tissues. Conventional radiotherapy has limitation because photons have exit dose to healthy tissues behind the tumor. In contrast, protons have Bragg peaks that give high radiation dose to the tumor but low exit dose or dose tail. Therefore, proton therapy is a promising candidate for the treatment of deep-seated tumors and tumors locating close to organs at risk. Moreover, the physical characteristic of protons is suitable for treating cancer in pediatric patients, especially primary brain tumor such as high-grade glioma (HGG) that has aggressive and high recurrent rates in pediatric patients. In this work, a computational approach was developed for calculating proton dose distribution from the pencil beam scanning method for treating primary brain tumor (HGG) in a pediatric patient using the Monte Carlo technique and the patient's anatomical data. Particle and Heavy Ion Transport Code System (PHITS) was used for calculating the proton dose distribution in the pediatric patient with a primary brain tumor. The CT-DICOM files were converted to the PHITS input file using the DICOM2PHITS program and the accuracy was tested against known values of electron densities in the Tissue Characterization Phantom GammexTM 467. A MATLAB-based proton therapy treatment planning program was used to create the beam delivery control file for the patient-specific Monte Carlo simulation. The treatment planning program required the patient's anatomical data, the depth dose distributions of monoenergetic (pencil beam) protons, the proton range-energy relationship, the stopping power ratios to water for protons in different materials and the scattering characteristics of protons in different materials. Most of these data were calculated using analytical formulas and their accuracy was tested against literature data or Monte Carlo simulations of monoenergetic protons carried out in this work, before the delivery control file was used by the Monte Carlo simulation. The Monte Carlo simulation gave a proton dose distribution output, which was compared with the proton dose distribution obtained from the MATLAB-based proton beam optimization program. The point dose verification technique was used for the comparison with a criterion of having dose difference in pixel-to-pixel of GTV of not exceeding 3% for at least 95% of the result. From this study, the DICOM2PHITS program was found to be suitable for converting CT-DICOM files to Monte Carlo inputs. The analytical functions used for calculating the proton beam data were found to be sufficiently accurate. The proton dose distribution comparison were not operate because some parameter was inaccurate from analytical function. This work was a pilot project for the development of a computational platform for dosimetric investigation of proton therapy using Monte Carlo simulation and patient CT data. The approach as carried out in this work could be interesting for researchers aiming to preliminarily investigate proton dose distribution in patients but do not have access to proton therapy machines and related equipment.