Optimization Of Implementation And Quality Assurance In Intensity Modulated Proton Therapy

Radiation therapy is one of the most widely used cancer treatments.It includes different modality radiation(such as X-rays,gamma,protons and heavy ions,etc.),which can kill cancer cell by themselves or in combination with surgery or chemotherapy.Due to the Bragg peak at the end of range in proton’s integrate depth dose curve,most of the proton energy can be accurately deposited in the tumour by modulate initial energy of proton.Compared to conventional photon radiation therapy,proton therapy,especially pencil beam scanning based Intensity Modulated Proton Therapy(IMPT)can generate a more conformal dose distribution to tumor and better protect healthy tissue.The therapeutic effect of proton radiotherapy strongly depends on the accuracy of treatment implementation,treatment implementation and QA is a complex work which is consist of many links,including treatment plan design,equipment and patient plan QA,patient positioning and image guidance,irradiation and QA after treatment.The implementation of each link will affect the treatment effect.Shanghai Advanced Proton Therapy facility(SAPT)is the first domestic proton radiotherapy facility independently developed under the cooperation of Shanghai Institute of Applied Physics,Shanghai Advanced Research Institute,Shanghai Jiaotong University Ruijin Hospital and Shanghai APACTRON Particle Equipemt Co.Ltd and now is in clinical trial stage.As a first domestic developed facility,in the process of equipment development and commissioning,researchers need to develop methods and technologies for design optimization and dosimetry verification according to the characteristic of the equipment itself,especially by evaluating and optimizing the impact of selected variables on the delivered dose distribution to optimize the treatment implementation and QA and improve the effect of treatment.Firstly,an independent dose calculation toolkit based on the pencil beam algorithm was developed,and did the experiments about the development and verification of this toolkit based on horizontal beamline data.The verification shows that: The maximum point-to-point absolute dose deviations between measurements and calculations was less than 3%,Gamma pass rates of relative plane dose distributions between measurements and calculations were better than 95% in 3%/3 mm.The precision of this independent dose calculation toolkit is acceptable for routine research requirements.Compared to commercial treatment planning system,the code of independent dose calculation toolkit can be modified.Thus,it can be used to investigate the impact of variables and as a tool for optimization of treatment implementation and QA in SAPT.Furthermore,during the commissioning of SAPT horizontal beamline,QA analysis toolkit based on logfile was developed by this independent dose calculation toolkit and achieved the fast evaluation to delivery.Secondly,combined with clinical requirements,based on developed independent dose calculation toolkit,specific issues during treatment implementation and QA,such as low resolution of two-dimensional(2D)ionization chamber(IC)array during QA before treatment,the patient set-up uncertainty during treatment,anatomic changes after treatment,etc.,were investigated on dosimetric aspect and proposed possible solutions to improve the therapeutic effect.During the machine commissioning and patient-specific QA before clinical treatment,we found out that if there are high dose gradient regions in the delivered dose distribution,due to the low resolution,the measurements of 2D IC array cannot reconstruct delivered dose distribution accurately.This can influence results of QA and then impacts the evaluation of delivered dose distribution.In this dissertation,we proposed to using Laplacian Pyramid Image Blending(LPIB)method to blend the low frequency part of measurement data of 2D IC array and high frequency part of recalculated dose distribution by independent dose calculation toolkit and obtain a reformatted dose distribution.Compared to measurement data,the reformatted dose distribution has high resolution and is more precise in high dose gradient region compared to direct interpolation of 2D IC array measurement.Before treatment,patients need set-up and verification(registration)by image guide system such as Cone Beam CT(CBCT).Different registration parameters such as Region Of Interest(ROI)and Degree Of Freedom(DOF)can influence the precision of the registration.In this dissertation,the impact of different registration parameters on precision of registration were evaluated from geometric,image and dosimetric aspects by independent dose calculation toolkit.Compared to geometric or image similarities,the dosimetric uncertainty can better evaluate the impact of parameters variation on registration and clinical effect.After several times fraction therapy,anatomy may change due to radiotherapy or dietary changes.In order to approaching optimal treatment,treatment plan should be adapted based on daily images.In this dissertation,based on independent dose calculation toolkit,we firstly proposed to using Water-Equivalent-Length(WEL)ratios between CT and CBCT to calibrate other CBCT and recalculated dose distribution based on calibrated CBCT.this calibration method can be used to online reoptimized plan based on CBCT and make the conformability of accumulated dose distribution in tumour better.The feasibility of this calibrating method was verified by a head and neck phantom and eight clinical cases.In conclusion,the treatment implementation and QA of IMPT were optimized,which can improve the treatment effect.