Font Size: a A A

Dose Distribution Study In Carbon Therapy

Posted on:2022-04-15Degree:MasterType:Thesis
Country:ChinaCandidate:C HuangFull Text:PDF
GTID:2504306491984849Subject:Nuclear Science and Technology
Abstract/Summary:PDF Full Text Request
Carbon ion beam has the advantages of definite range and sharp high dose Bragg peak,which can deposit precise dose in tumors and reduce toxic effects in normal tissues.Therefore,carbon therapy technology has been developed rapidly.Carbon ion radiography can provide the relative blocking coefficient distribution of carbon ions in patients before therapy,which is beneficial to ensure the accuracy of dose deposition.Compton imaging is one of the methods for real-time monitoring of the 3D dose distribution during carbon therapy,which can monitor the accuracy of the treatment dose distribution.This thesis explores the dose distribution monitoring before and during carbon therapy.In the research of carbon ion radiography,a combined array ionization chamber detection system was designed,which achieves radiographic imaging and 3D dose measurement.The detection system measures the depth dose curve and the lateral spread of the beam spot,and realizes dynamic imaging in raster scan or point scan mode.Based on the parameters of the heavy ion medical machine(HIMM)in Lanzhou,we simulated 8 hemisphere phantoms with radii of 20 mm to 60 mm in GATE,and developed a neural network algorithm which combines the support vector machine(SVM)and the backpropagation(BP).The trained neural network was used to reconstruct the unknown phantoms with hemisphere,step and slope shapes.Results showed that the prediction accuracy of a 40 mm hemisphere phantom is 98.64%.The study of step phantom and slope phantom proved the generalization performance of the neural network algorithm.In the research of carbon ion Compton imaging,we analyzed prompt gamma-ray detection and the image reconstruction.A relationship between prompt gamma-ray 3D location and dose distribution was established.We optimized imaging detector parameters using MC simulation.A 20×20×25 mm3 CLYC crystal and a 60×60×60 mm3 CZT crystal were selected as the scattering detector and absorbing detector of Compton imaging system,respectively.We simulated the prompt gamma-ray generation when carbon ion beam irradiated a PMMA phantom.A 9.3%deviation was found between the reconstructed dose distribution and the real dose distribution.To verify the feasibility of Compton imaging method,we used four LYSO detector pairs with 1.5mm pitch to image a 22Na point source with 3 mm in diameter.The LYSO Compton imaging system obtained an image of 4.05 mm FWHM.The size from point source to the reconstructed Compton image was enlarged 1.35 times.
Keywords/Search Tags:Carbon ion therapy, Dose monitoring, Neural network algorithm, Support vector machine(SVM) network, Backpropagation(BP) network, Compton imaging, Ionization chamber detector, CLYC crystal, CZT crystal
PDF Full Text Request
Related items