de Boer R W
Radiotherapy Department, The Netherlands Cancer Institute, (Antoni van Leeuwenhoekhuis), Amsterdam.
Radiother Oncol. 1988 Apr;11(4):361-7. doi: 10.1016/0167-8140(88)90207-1.
In studies of radiation damage, the parameter alpha/beta of the linear-quadratic (LQ) model is usually estimated from iso-effect data by plotting the iso-effect doses in a so-called Fe-plot, which is a plot of the inverse total dose versus dose per fraction. This method, however, is statistically not well founded. A better, but relatively complex technique to estimate the alpha/beta ratio was described by Tucker. The present contribution shows that results, virtually identical to the ones of Tucker can be obtained in a very simple way if the iso-effect doses are plotted as total dose Dn versus the product of dose per fraction and total dose, Dndn. A reliable estimator of the alpha/beta ratio consists then of the negative inverse slope of the linear-regression line in this plot. The presentation of iso-effect data in such a D-dD plot highlights deviations from the LQ model more clearly than does the Fe-plot. It is therefore suggested to use the D-dD plot as a quick and easy graphical scheme to obtain an estimate of alpha/beta.
在辐射损伤研究中，线性二次（LQ）模型的α/β参数通常通过在所谓的Fe图中绘制等效应剂量来从等效应数据中估计，Fe图是总剂量倒数与分次剂量的关系图。然而，这种方法在统计学上没有充分的依据。Tucker描述了一种更好但相对复杂的估计α/β比值的技术。本文表明，如果将等效应剂量绘制为总剂量Dn与分次剂量和总剂量的乘积Dndn的关系图，那么可以用一种非常简单的方法获得与Tucker的结果几乎相同的结果。在该图中，α/β比值的可靠估计值由线性回归线的负倒数斜率组成。与Fe图相比，在这种D-dD图中呈现等效应数据能更清楚地突出与LQ模型的偏差。因此，建议使用D-dD图作为一种快速简便的图形方案来获得α/β的估计值。
