透過您的圖書館登入
IP:3.16.147.124
  • 學位論文

簡化型蒙地卡羅射源模型 應用於筆型射束掃描質子治療之劑量計算研究

A Simplified Monte Carlo Source Model used in Dose Calculation for Proton Pencil-beam Scanning

指導教授 : 許榮鈞 陳俊丞
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


本研究發展一套簡化的蒙地卡羅射源模型適用於筆型射束掃描式質子治療的三維劑量分佈計算,兼具合理的計算效率與劑量準確度,是一種除了治療計劃系統外的獨立劑量驗證方法。 由於筆型射束掃描式質子治療的特性,其射源項的模擬可省去許多傳統質子治療機頭內複雜組件的影響,本研究的射源模型考慮了與質子射束特性直接相關的重要參數,這些參數的來源有二:第一部分的參數來自臨床治療前試運轉的測量結果,藉由測量不同能量(100-226 MeV)質子束在水箱假體的積分深度劑量分佈,焦斑大小及有效射源位置,可以決定一簡化之射源項;第二部分包括依計畫腫瘤體積與位置選定之射束能量、焦斑位置與監控單位。 本研究依射束特性的測量結果建立了一個簡化型的射源資料庫,配合直接來自治療計劃系統輸出檔案中的治療參數,自動整合至蒙地卡羅程式FLUKA進行模擬,該蒙地卡羅計算劑量結果可透過適當參數修正而成為三維空間絕對劑量分佈,再與治療計劃劑量計算結果與實際測量結果進行比較。 本研究選用了一系列不同案例來檢驗此簡化射源模型是否足夠完善,從最簡單的矩形模型到實際臨床攝護腺案例。計算時間若使用本實驗室24核心的電腦,每個案例的計算都能夠在一小時內完成,平均計算所需時間約為四十分鐘,治療計劃腫瘤體積內計算結果的統計誤差都在2%以內,且與治療計畫計算結果或實際量測相比都有很高的一致性。假體中不同深度下的二維劑量加馬評估通過率幾乎都在95%以上,證明此簡化型蒙地卡羅射源模型確實能夠準確地預估筆型射束掃描式質子治療的三維劑量分佈,並兼具合理的計算效率。未來持續的發展應可用於例行品保作業或臨床治療計劃的獨立劑量驗證工具。

並列摘要


A simplified Monte Carlo (MC) source model for proton pencil-beam scanning (PBS) was proposed and demonstrated in this study. Three-dimensional dose distribution in target can be obtained by feeding treatment field parameters exported from a treatment planning system (TPS) into the source model. A PBS source term (initial energy and spread) as a function of proton energy (100-226 MeV) was constructed to directly imitate the measured integral depth-dose curves (IDDCs), which is independent of any component in treatment nozzle. The spot divergence and deflection were determined from the measured spot sizes at the isocenteric plane and source-to-axis distance respectively, and the absolute dose per particle at 2-cm depth for 100-226 MeV protons were used as the MU weighting correction factors. Simple geometry structures and TG-119 phantoms were created in the TPS as the planning target volumes (PTVs) in a 40×40×40 cm3 water phantom. The range shifters (water equivalent thickness of 4.0 cm and 7.5 cm) were used for the PTVs at depths <10 cm. The calculated depth-dose curves, point doses, and plane dose distributions by the simplified MC source model were compared with measurements using multi-layer ion chambers, parallel plate ion chamber, and MatriXX PT detectors (IBA dosimetry GmbH, Schwarzenbruck, Germany), as well as the TPS. The comparisons of absolute dose profiles at the isocenter depth were with the error within 2% compared in plateau region. The Gamma passing rate were showed and almost all of cases can reach the goal of 95% compared with the TPS calculations and measurements respectively, which met the criteria of 3% and 3 mm in a interpolated grid size of 1 mm. Over all, the typical PBS treatment field simulations could be completed in an hour using a 24-core computing node. The simplified Monte Carlo source model provides accurate 3D dose distribution in water with clinically-acceptable computational efficiency. The current model could be used for the commissioning of treatment planning system, independent MU check, and patient-specific quality assurance in a homogeneous water phantom.

參考文獻


[1] M. Goitein and T. Miller, "Planning proton therapy of the eye," Med. Phys., vol. 10, pp. 275-283, 1983.
[2] A. Tourovsky, A. J. Lomax, U. Schneider, and Pedroni, "Monte Carlo dose calculations for spot scanned proton therapy," Phys. Med. Biol., vol. 50, pp. 971-981, 2005.
[3] L. Hong, M. Goitein, M. Bucciolini, R. Comiskey, B. Gottschalk, S. Rosenthal, et al., "A pencil beam algorithm for proton dose calculations," Phys. Med. Biol., vol. 41, pp. 1305-1330, 1996.
[4] N. Koch and W. Newhauser, "VIRTUAL COMMISSIONING OF A TREATMENT PLANNING SYSTEM FOR PROTON THERAPY OF OCULAR CANCERS," Radiation Protection Dosimetry, vol. 115, pp. 159-163, 2005.
[5] C. Rethfeldt, H. Fuchs, and K.-U. Gardey, "Dose distributions of a proton beam for eye tumor therapy: Hybrid pencil-beam raytracing calculations," Med. Phys., vol. 33, pp. 782-791, 2006.

延伸閱讀