Nonparametric ROC and LROC analysis
Nonparametric ROC and LROC analysis
复制标题
DOI：
10.1118/1.2717407
复制
发表时间：
2007-05-01
期刊：
影响因子：
3.8
通讯作者：
Popescu, Lucretiu M.
中科院分区：
文献类型：
--
作者：
Popescu, Lucretiu M.
在本文中，我们回顾了非参数受试者工作特征（ROC）分析的几个结果，并提出了一个扩展的非参数局部化ROC（LROC）分析。方程的特征曲线下的面积的估计和方差计算。还提出了信号缺失和信号存在的图像样本的数量之间的最佳比率的选择的表达式。结果可以应用于连续或离散的评分尺度。仿真研究验证了理论推导的正确性，并表明LROC分析比ROC分析更为灵敏。(c)2007年美国医学物理学家协会。
In this paper we review several results of the nonparametric receiver operating characteristic (ROC) analysis and present an extension to the nonparametric localization ROC (LROC) analysis. Equations for the estimation of the area under the characteristic curve and for the variance calculations are derived. Expressions for the choice of the optimal ratio between the number of signal-absent and signal-present image samples are also presented. The results can be applied both with continuous or discrete scoring scales. The simulation studies carried out validate the theoretical derivations and show that the LROC analysis is considerably more sensitive than the ROC analysis. (c) 2007 American Association of Physicists in Medicine.