| Radiation therapy is one of the main means of treating malignant tumors at present,and the design of radiation treatment plan depends on the treatment planning system.The dose calculation is the core of the treatment planning system,and the accuracy of the dose calculation will directly affect the final treatment effect.Collapsed cone convolution(CCC)algorithm is a model-based analytical dose calculation algorithm.To ensure the accuracy of dose calculation,the beam modeling process is necessary,and the modeling of photon energy spectrum is an important part.The traditional modeling method of the photon energy spectrum for the CCC algorithm is based on manual adjustment.This modeling method is highly dependent on the modeling experience of the modeler,and it is difficult to find the optimal model within the limited modeling time.Aiming at the limitations of traditional manual modeling methods,the research purpose of this thesis is to design and develop a photon energy spectrum automatic modeling program oriented to CCC algorithm,to realize the automation of photon energy spectrum modeling process,and to use the treatment planning system KylinRay-TPSe as platform to verify the accuracy of the program and test the feasibility of clinical application.Based on this research purpose,the research content of this thesis is divided into the following two aspects:(1)Implementation and accuracy verification of the photon energy spectrum automatic modeling program oriented to the CCC algorithm:The genetic algorithm was introduced into the photon energy spectrum modeling process of the CCC algorithm,and the fitness function of the genetic algorithm was constructed based on the mean square error of percentage depth dose curve between the calculation value of the CCC algorithm and the experimental measurement value.By partitioning the percentage depth dose curve and setting the weights for the calculation errors in each area,the optimization of the key areas of the photon energy spectrum was realized.Based on the above automatic modeling method of photon energy spectrum,a photon energy spectrum automatic modeling program oriented to CCC algorithm was developed,and the accuracy of the program was verified based on the cases in the YY 0775 national industry standard using KylinRay-TPSe as a platform.The verification results showed that the calculation results of the CCC algorithm using the photon energy spectrum automatic modeling program passed the YY 0775 standard,which proves that the photon energy spectrum automatic modeling program has high accuracy.(2)The clinical application test of the photon energy spectrum automatic modeling program oriented to the CCC algorithm:The cases stipulated in the YY/T 0889 national industry standard and some real clinical patients(including 5 patients with tumors in head and neck,in chest and abdomen,and in pelvic cavity each)were used as test cases.We used the CCC dose calculation engine of KylinRay-TPSe to calculate the dose of all test cases based on the photon energy spectrum obtained by the automatic modeling program.For the YY/T 0889 standard cases,the dose calculation values at the specified position were compared with the actual measured values.For the clinical cases,the calculation results of KylinRay-TPSe and Pinnacle3 under the same plan were compared in the difference of dosimetric parameters in the planning target volume and organs at risk,and the calculation results of the two treatment planning systems were compared using gamma analysis in the PTW VeriSoft software.The test results based on the YY/T 0889 standard showed that the calculation results of all test cases can meet the standard requirements,which proves that the program has high accuracy for intensity modulated planning.The test results of clinical cases showed there was no statistically significant difference between the calculation results of the CCC algorithm using the photon energy spectrum automatic modeling program and the calculation results of Pinnacle3 in the dosimetry parameters of the planning target volume and organs at risk(P>0.05).Under the contrast condition of±3%/3mm,the 2D gamma pass rate of the isocenter plane was above 95.1%,and the 3D gamma pass rate was above 95.0%,which proves that the accuracy of the calculation results of the CCC algorithm using the photon energy spectrum automatic modeling program has reached the level of mainstream commercial treatment planning system,and the program can meet the requirements of clinical use. |