在放射腫瘤治療部門，使用電腦斷層影像來判斷腫瘤解剖位置及正確的照野方向，影像品質也更加受到重 視，因此必須針對影像品質做出適當規範。程序書並沒有對影像品質的操作方式有明確的規定，希望藉由機械測 量的穩定性來證明專家評分方法是可行的。為確保電腦斷層模擬定位儀影像的空間解析度，本研究利用田口方法 找出電腦斷層模擬定位儀最佳化的掃描參數設定。本實驗使用Light Speed 電腦斷層掃描儀、電腦斷層模擬定位 儀專用校正假體、掃描對比細分元件。由五位放射腫瘤科部門人員針對影像細分之解析度及影像品質評分，找出 最佳化組合參數。為避免專家評分方法因人為不客觀等因素，導致研究結果的可信度降低，因此再以電腦斷層影 像處理軟體ROI(Region of Interest)測量對比細CT 值，來驗證專家評分方法的可信度。經由變異分析後，兩者最 佳化組合參數為管電壓120 kVp、管電流300 mA、掃描照野20 cm、掃描厚度5 mm 為最佳化組合參數。本研究 經由田口方法找出電腦斷層模擬攝影機最佳化組合參數，結果證明兩者有相同的最佳化組合參數，驗證專家評分 是可信賴的方法。

Radiation Oncology department was used CT Imaging to determine the tumor location and correct irradiation direction as well as the irradiation field. There is also more attention and adequate standard must be put for image quality. Program book does not have clear rules for image quality operation. We hope to use mechanical measurements to prove the stability of an expert rating method is feasible. In order to ensure the spatial resolution of CT simulation location image, this study using Taguchi method to optimize the computer simulation of scanning parameters. In order to find the optimal combination of parameters, this experiment using Light Speed computer tomography simulation, CT specific correction phantom, scanning contrast segments of image, and five diagnostic Radiologist for scoring image resolution and image contrast of the segmentation quality. To avoid reducing the credibility of the results from experts scores due to human errors, we choose computed tomographic image processing software circle ROI (Region of Interest) and measuring fine contrast CT values to verify the credibility of expert evaluation methods. Results data via ANOVA analysis show the optimal combination of parameters for the tube voltage is 120 kVp, tube current is 300 mA, scanning FOV (field of view) is 20 cm, and the scanning thickness is 5 mm. This study use Taguchi method to find the optimized parameters of CT simulation. The results show that both have the same combination of parameters, and verify that the expert scoring is also reliable.