Fierz Walter
labormedizinisches zentrum Dr Risch, Vaduz, Liechtenstein.
PLoS One. 2018 Feb 22;13(2):e0192420. doi: 10.1371/journal.pone.0192420. eCollection 2018.
Receiver operating characteristic (ROC) analysis is widely used to describe the discriminatory power of a diagnostic test to differentiate between populations having or not having a specific disease, using a dichotomous threshold. In this way, positive and negative likelihood ratios (LR+ and LR-) can be calculated to be used in Bayes' way of estimating disease probabilities. Similarly, LRs can be calculated for certain ranges of test results. However, since many diagnostic tests are of quantitative nature, it would be desirable to estimate LRs for each quantitative result. These LRs are equal to the slope of the tangent to the ROC curve at the corresponding point. Since the exact distribution of test results in diseased and non-diseased people is often not known, the calculation of such LRs for quantitative test results is not straightforward. Here, a simple distribution-independent method is described to reach this goal using Bézier curves that are defined by tangents to a curve. The use of such a method would help in standardizing quantitative test results, which are not always comparable between different test providers, by reporting them as LRs for a specific diagnosis, in addition to, or instead of, quantities such as mg/L or nmol/L, or even indices or units.
受试者工作特征（ROC）分析被广泛用于描述诊断测试区分患有或未患有特定疾病人群的鉴别能力，使用二分阈值。通过这种方式，可以计算阳性和阴性似然比（LR+和LR-），以便用于贝叶斯估计疾病概率的方法。同样，可以针对测试结果的特定范围计算似然比。然而，由于许多诊断测试具有定量性质，期望为每个定量结果估计似然比。这些似然比等于ROC曲线在对应点处切线的斜率。由于患病和未患病个体中测试结果的确切分布通常未知，因此针对定量测试结果计算此类似然比并非易事。在此，描述了一种简单的与分布无关的方法，使用由曲线切线定义的贝塞尔曲线来实现这一目标。使用这种方法将有助于标准化定量测试结果，不同测试提供者之间的结果往往不可比，除了报告诸如mg/L或nmol/L等数量，甚至指数或单位之外，还可以将它们报告为特定诊断的似然比，或者取而代之。
