Dosimetry Study Of Multicriterion Optimization Algorithm For The Nasopharygeal Carcinoma Radiotherapy

Radiotherapy, together with the surgery and chemotherapy are the threemain means for tumor treatment. Among them,more than70%patients needRadiotherpy.Rencently,intensity modulated radiotherpy (IMRT) has beenwidely used to clinical oncology as an advanced technique which modulatingsegments for all beams in order to get a better tradeoff between tumor andorgan at risk(OAR).As the most representative of IMRT, nasopharyngealcarcinoma (NPC) radiotherpy need consider more objectives,which has greatsignificance to study the dosimetry.The IMRT of NPC is an inverse optimization which is a typicalmulti-objective problem.However,most treatment planning system (TPS) dealwith it as a single-objective,who dose not satisfy the multi-objectivecharacteristic of IMRT inverse planning. Multi-objective optimizationoriginated from designing、modeling and planning in the application ofactual system. This thesis study multi-objective optimization in theapplication of IMRT for NPC,which solved the following problems:For Intsentiy modulated radiation therapy, The present article not onlytake into account the effect of the tumor, but also consider the sparing ofnormal organs, we offer a equivalent uniform dose(EUD) function asmulti-objective optimization model with characterize the target dose coverageand normal organ dose limit. The method by theoretical analysis and clinical applications, the results show that the objective function characterizationusing EUD model has good convexity, easy to find multi-objective Paretooptimal solution, compared with the previous modeling method based onphysical dose in clinical applications patients can reduce the radiation dose tonormal organs.In this paper, the improved multi-objective optimization method isapplied to nasopharyngeal carcinoma model, the model is studied test resultsshow that the optimization method can not only improve the speed of NPCoptimization, and improved accuracy can be Pareto optimal solution.Lastly, we use the dose volume histograms（DVH）、normal tissuecomplication probability（NTCP） assessment tools for all planning, the useof the multi-objective optimization into the dosimetry results obtainednasopharyngeal carcinoma, and dose and single-objective optimizationmethods and existing multi-objective optimization method comparison of theresults of science, and the results showed that: after multi-objectiveoptimization methods to improve patient target conformal index and evennessindex is superior, normal organs are lower radiation dose, reduced risk ofnormal tissue complication probability...