| Cancer is a major cause of increased mortality worldwide,and radiation therapy can be adapted to treat aggressive tumors.Glioma is one of the most life-threatening and common brain tumors.It originates from nerve stroma cells and has the ability to rapidly infiltrate surrounding healthy tissue.Although less radiosensitive,radiation therapy is still the most effective treatment other than surgery.Microdosimetry can provide a quantitative description of the randomness of energy deposition in small volumes.In the case that it is difficult to obtain the absorbed dose of a single cell or a community of cells during radiotherapy,Monte Carlo software is usually used as a means to obtain theoretical values because it can simulate the interaction between radiation and substances.Microdosimetry index line energy can be connected with the theory of dual radiation effect,and it is of great significance for radiation therapy and radiation protection to carry out y prediction of the mean value of site line energy of different diameters at the microscopic scale(0-1000nm)concerned with biological effects.Cells are the most basic unit in the organism.Different types of tumor cells absorb different doses during irradiation.When radiation interacts with glioma cells,its physical and biological processes become more complex.The purpose of this study is to explore the construction method of biology-physical cell model for the application of radiotherapy and exploring the mechanism of biological effects by taking glioma cells as the research object,combining radiation physics technology and biological experimental technology.In terms of research methods and contents,the main work of this paper is as follows:using radiation microdosimetry theory,combining advanced image analysis and numerical analysis technology,the real glioma cell surface model is constructed,monte Carlo simulation and radiobiology experiments are carried out,and the comparative experimental results of radiation physics and radiobiology are obtained.In this paper,the neural network prediction model for the mean line energy dose of sites with different diameters was constructed based on the line energy mean database of microdosimetric index calculated by low-energy electrons in liquid water using Monte Carlo software GEant4-DNA toolkit.The BP neural network model was optimized by MEA algorithm compiled by MATLAB platform.According to the empirical formula,the hidden layer and neuron number of neural network were adjusted,the iterative step number and training algorithm were set,and the mean line energy of single energy electrons and spectral distributed electrons in the database was predicted.The relative error range between the prediction results verified by the model and monte Carlo calculation results is 0.03%-5.98%(the error range is 0.1%-5.98%for single energy electrons and 0.03%-4.4%for spectral distributed electrons),showing a good prediction accuracy.The results show that the MEA-BP neural network is the best solution considering the prediction accuracy and operation time.In this paper,a method to construct glioma cell surface model is proposed for the first time.Cellular curved models based on dose distribution,biological parameters,and medical images may help to improve radiotherapy modalities and optimize treatment plans.By fluorescent staining of cytoskeleton and nucleus,3D layer description of cells was carried out by Leica laser confocal to obtain tif images.The initial images and single cell images with good scanning were cut by MATLAB self-programmed program,and the image format was transformed into DICOM images with coordinate location information.Amira and Meshmixer image processing software were used to repair the incomplete parts of the DICOM image and generate a surface model.Spekpy software was used to simulate the same 160kV energy spectrum of X-ray machine in the irradiation center,so as to achieve a more realistic irradiation effect.Monte Carlo software GATE was used to simulate negative X-ray irradiation along the z-axis of cells,and dose distribution was estimated for single cells and community cells respectively.According to the statistics of the number of photons generated by X-ray machine simulation,the dose rate of X-ray to cytoplasm is 48.4mGy/s,and that of nucleus is 14.5mGy/s.At the irradiation time,the nuclear deposition dose of a single cell is about 70%of the external irradiation dose.As 2Gy is the general dose in the radiotherapy segmentation scheme,the average dose of 2Gy was calculated for the dose distribution of the nucleus and cytoplasm in the community cells,and it was found that the dose distribution inside the community cells presented a Gaussian distribution,reflecting the randomness of dose deposition,in which the nuclear dose was concentrated in 0.6-1.8Gy.Cytoplasmic doses were concentrated between 0.9-2.4Gy.In addition,combined with monte carlo dose estimation results,the mechanism of radiobiological effects was explored,and biological phenomena of proliferation and apoptosis of glioma cells induced by different doses of X-ray irradiation(1,2,4,6,8Gy)were tested by biological means.The results of radiobiology experiment and physical simulation dose distribution are discussed.As the dose increased,it was found that foci number decreased by EdU on DNA labeling,and the proliferation ability was weakened by plate cloning.The results of immunofluorescence assay showed that ionizing radiation caused DNA double bond breakage,the expression of phosphorylated H2AX increased gradually,and foci number increased.Combined with dose distribution estimation of community cells and flow cytometry analysis,under 2Gy irradiation,the nuclear deposition dose of community cells was mainly distributed within 0.6-1.8Gy.With the lapse of time after irradiation,apoptosis rate of glioma cells increased and reached the highest 31.7%at 48h(early apoptosis rate was 21.5%,late apoptosis rate was 10.2%).After 72h,the apoptosis rate was only 25.82%(early apoptosis rate was 4.02%,late apoptosis rate was 21.8%),and the gradual decrease of apoptosis rate indicated that the cells began to repair.The results of this study will play a positive role in expanding the application of microdosimetry and Monte Carlo simulation in the process of radiation therapy for glioma. |
PDF Full Text Request