蒙地卡羅模擬直線加速器輸出劑量與各模組後之通量研究

Study of Monte Carlo Simulation on Output Dosimetry and Fluence after Each Component Module of LINAC

張國平(Kwo-Ping Chang) ； 王誌瑋(Chi-Way Wang) ； 陳佳君(Cha-Chun Chen) ； 蕭安成(An-Cheng Shiau) ； 張席昌(Hsi-Chang Chang) ； 劉文山(Wen-Shan Liu)

《中華放射線醫學雜誌》 32卷4期 (2007/12) Pp. 185-191

摘要

由於不均勻介質會影響放射治療計畫系統(treatment planning system, TPS)的劑量計算而可能造成誤差，因此本研究利用蒙地卡羅模擬與實驗以及治療計畫之比對來探討這些影響的關係。首先是正確的模擬直線加速器（Varian 21EX, 6MV光子模式）各模組、幾何圖形與入射電子能量與寬度；當蒙地卡羅模擬值可以與標準水假體的實測值一致時，進行各模組後之通量變化之探討。接著，利用簡單的固態圓柱型假體(cylinder phantom)、胸腔治療病人的治療計畫系統，在不同方向的射束下，加上CT掃描的資料（此配合程式所需建立的CT-phantom），然後比對TPS計算值
與蒙地卡羅計算值。結果顯示，模擬結果之最佳參數為(E=6.14 MeV; FWHM=0.03cm)，以10×10平方公分與40×40平方公分的照野模擬百分深度劑量(PDD)與劑量剖面(dose profile)，游離腔實驗值與蒙地卡羅模擬值非常吻合。整平濾片之後，粒子的通量與能量通量均有很大的改變。而在二次準直儀之後的通量中央較兩邊為低，能量通量則很平均，因為中央光子的能量稍高於兩邊非中央的部分。針對圓柱型假體，我們發現蒙地卡羅模擬值與TPS所得到的等劑量曲線(isodose curve)很吻合，也證實了蒙地卡羅方法模擬不同射束方向的可行性。在病人劑量方面，TPS計算的等劑量圖與蒙地卡羅模擬計算的等劑量圖，結果發現除了某些界面部分，TPS計算值與蒙地卡羅模擬值很接近。

並列摘要

Due to the possible errors resulted from TPS (treatment planning system) dose calculations for inhomogeneous media, Monte Carlo simulation was used in this study and compared to the data of the measurements and those of TPS. First, we need to accurately simulate the modules and geometry of LINAC (Varian 21EX, 6MV photon mode) as well as the energy/width of the incidence electron. After the data from MC simulation and measurements have been matched, the fluence variances of each component module were studied. The comparisons between the isodose curves calculated from TPS and Monte Carlo method have been performed for two cases, the cylinder phantom and the patient, where the CT images were used both for TPS and MC simulation. Under the optimal parameters (E=6.14 MeV；FWHM=0.03 cm) found in this study, the PDD (percentage depth dose) and dose profile for 10×10 cm^2 and 40 × 40 cm^2 field sizes of Monte Carlo simulation agreed with the measurements of the ionization chamber. It was found that the fluence and energy fluence changed enormously below the flattening filter. Less fluence in the central region was found below the secondary collimator; however, because the energy of the photons in the central region is slightly higher, the energy fluence was found to be equally distributed below the secondary collimator. For the cylinder phantom, the isodose curves calculated from Monte Carlo simulation agreed with those calculated from TPS, such results also verified the feasibility of Monte Carlo simulation for different directions of the beam. For the patient, it was found that the isodose curves were also matched between the data of TPS and Monte Carlo simulation, except for some regions of interfaces.