| Dose calculation algorithm is paid attention to because it is one of the core modules on radiotherapy treatment planning system (TPS/RTS). There is important signification for developing inverse planning system in intensity modulated radiation therapy (IMRT) with seeking a high-speed and precise dose calculation algorithm. The dose calculation algorithm is the main discussion topic in this paper. Concretely, three parts are included as follows:Monte Carlo (MC) method is recognized as the most powerful tool for simulating particle transport, but it is limited in clinical TPS because of the fatal shortcomings, such as the slow speed of convergence and the too long computation time. Therefore, this project introduce 3D filtering techniques for accelerating MC simulations, improve the 3D Gauss and Savitzky-Golay filters, and put forward the parallel and cascade mixture filters. Error analyses indicated that the 3D filters can effectively smooth MC rough dose distributions.Based on the previous work of the group for FsPB (Finite-size Pencil Beam) algorithm of matrix model based on MC simulation, this work redesign the algorithm flow and reduce the order of complexity so that the computation speed is been effectively accelerated. Under the condition of water phantom, test results show that the redesigned algorithm increases the computation speed several decuple times without decreasing the precision.In order to get the dose distributions in human tissue, the result computed by matrix model FsPB algorithm in water phantom should be corrected to the dose in human body. This work gets the tissue inhomogeneity information from CT images. Firstly, the algorithm transform the CT values into electron density, and then compute the correction factors using 1D Batho algorithm, finally get the dose in human body. The clinical test example indicates that the Bath algorithm can satisfy the requirement of clinical application in most cases with the high speed.
|
PDF Full Text Request