使用田口方法找出提升核醫科電腦斷層影像解析度之最佳化條件

Using the Taguchi method to determine the optimization factor for improving nuclear medicine computed tomography images

張雅蓮(Ya-Lien Chang) ； 楊淑琴(Shu-Qin Yang) ； 何聖佑(Sheng-Yow Ho)

《醫療品質雜誌》 14卷4期 (2020/07) Pp. 54-59

https://doi.org/10.3966/199457952020071404011

摘要

利用田口方法L9（3^4）模擬腹部電腦斷層影像品質條件，並配合校正專用壓克力水假體，選定以下4個因子來作為最佳化田口實驗設計法的變動因子：A.管電壓（kVp），B.管電流（mA），螺距比（Pitch），D.掃描厚度（Thickness）。利用田口實驗設計法中的L9（34）直交表，依此設定掃描參數的4個控制因子、3個變動水準、執行9次實驗組合因子水準排列組合，共可得9張對比細分影像，並依此系列影像得到ROI之測量平均值及標準差、CTDI輻射劑量，並利用統計方法產生ROI測量法之因子反應表與反應圖。由ROI測量法S/N因子反應圖得知最佳化因子組合為A3 B3 C1 D3，而y8實驗組合A3 B2 C1 D3僅是此次實驗中最佳的組合表現（Local Optimal Parameters），因此A3 B3 C1 D3才是最佳參數組合（Global Optimal Parameter Design）。意謂最佳化影像特性之電腦斷層參數設定為管電壓140kVp、管電流120mA、螺距比0.938mm、掃描厚度5mm。

關鍵字

電腦斷層掃描儀 ； 田口法 ； 最佳化因子

並列摘要

This study used the L9(3^4) orthogonal array in the Taguchi experimental design method and evaluated the four control factors of the scanning parameters, the three levels of variation, and the nine experimental combinations of the level combinations. A total of nine comparative subdivision images were obtained. The average and standard deviation of the region of interest (ROI) were obtained from the images. Then, the correct combination of optimization factors was determined using a factor response graph. F-Check in the variation analysis table was used to determine the significance of each factor, based on its p value, as a basis for assessing the combination of optimization factors. The S/N ratio factor reaction diagram for ROI measurement indicated that the optimization factor combination was A3B3C1D3, which suggests that the optimum imaging conditions are a tube voltage of 140 kVp, a tube current of 120 mA, a pitch ratio of 0.938, and a scanning thickness of 5 mm.

並列關鍵字

Computed tomography ； Taguchi experimental design method ； Optimal factor