目的：本研究的目的在評估SmartArc內不同計算參數設定對弧形治療準確度的影響， 我們希望能由數個不同患部的治療計畫中，分別以不同旋轉機頭角度間距還算及是否考量治療床衰減等，實際量測治療之準確度差異。材料與方法：共選擇28例弧形治療病患，其中包含8例腹部／骨盆、6例單側頭頸部、8例雙側頭頸部及6例攝護腺。使用E1ekta Synergy直線加速器及Philips Pinnacle 9.0之SmartArc運算程式執行治療及還算，分別以旋轉機頭角度間距2、3、4及6度執行最佳化計算，並於每組影像加上寬50 cm厚8 mm密度1g／立方公分的模擬治療床。照射時，假體內同時擺放0.125立方公分Semif1ex游離腔及Gafchromic EBT2底片。結果：於所有224個治療計畫中考量有／無治療床差異，符合點劑量±3%基準內之比率分別為99.1%及94.6%；EBT2底片絕對劑量分佈於3%/3 mm的加碼指標通過率，低於95%閾值比率分別為5.4%及11.6% 。針對臨床使用參數設定條件（角度間距3 度、考量治療床衰減），腹部/骨盆、雙側頭頸部、雙側頭頸部、攝護腺於3% / 3 mm的加碼指標通過率分別為98.3±2.3%、99.5±0.5%、97.6±2.3%及99.5±0.5%。Kruskal-Wallis H檢定顯示2-6度角度間距之劑量實測結果，並無顯著差異且未達要求案例隨機發生在任何角度間距；Mann-Whitney U檢定顯示，攝護腺以外之其他患部在有／無治療床之p值低於或接近顯著水準。Spearman's correlation coefficient更指出照射MU量可能影響治療準確度（p＜0.05）但與PTV大小則無相關性。結論：利用EBT2底片執行弧形治療前之整體劑量評估確為一可行方法，本研究結果顯示，雖然在不同計算參數設定下治療準確度差異不一，但整體來說，其劑量測量值與計算值之一致性極高。

Purpose: The goal of this study was to investigate the accuracy of VMAT with different calculation parameters in SmartArc. Dosimetric comparisons of various gantry spacings with and without couch (WC/WOC) insert simulation were evaluated based on statistics.Materials and Method: Twenty-eight VMAT cases were selected in this study, including 8 abdomen/pelvis, 6 one-side HN, 8 two-side HN and 6 prostate. A commercial Elekta Synergy linear accelerator and SmartArc algorithm in Pinnacle^3 9.0 was used. The patients were treated clinically with VMAT using 3° gantry spacing and reoptimized using 2°, 4° and 6°. An 8 mm thick water equivalent contour couch was added under the phantom. The phantom was irradiated with 0.125 cm^3 Semiflex chamber and Gafchromic EBT2 film simultaneously.Results: 99.1% and 94.6% of WC and WOC passed the 3% point dose difference criteria. In absolute dosimetric agreement of EBT2 film, the gamma passing rate of 3%/3 mm for all 224 plans separately were 5.4% for WC and 11.6% for WOC below 95% tolerance level. While using clinical parameters (3° gantry spacing, WC), the 3%/3 mm gamma average passing rate was 98.3±2.3%, 99.5±0.5%, 97.6±2.8% and 99.5±0.5% for abdomen/pelvis, one-side HN, two-side HN and prostate respectively Considered with the 2-6° gantry spacing, the Kruskal-Wallis H test shows very similar results and failures situations occurred randomly on any gantry spacing. The results 01 with /without couch insert simulation shows the Mann-Whitney U test p-value lower than or close to, the significant level expected for prostate cases. Spearman's correlation coefficient indicates that delivery MU may inlluence the accuracy (p-value < 0.05) but it's independent of PTV volume.Conclusions: Composite dose measurements performed with EBT2 film are a practicable approach for patient-specific QA of VMAT plans. Film dosimetry in this study has shown good agreement in various parameter settings.