MAXIMUM-LIKELIHOOD ESTIMATORS OF SENSITIVITY AND SPECIFICITY CORRECTED FOR VERIFICATION BIAS
MAXIMUM-LIKELIHOOD ESTIMATORS OF SENSITIVITY AND SPECIFICITY CORRECTED FOR VERIFICATION BIAS
复制标题
DOI：
10.1080/03610929308831209
复制
发表时间：
1993-01-01
影响因子：
0.8
通讯作者：
ZHOU, XH
中科院分区：
文献类型：
--
作者：
ZHOU, XH
灵敏度和特异性是诊断测试准确性的两种常用测量方法。要计算这两个指标，必须知道每个患者的真实疾病状态。然而，在实践中，有时一些具有测试结果的患者不被选择用于验证疾病状态。如果计算中仅使用已验证疾病状态的患者，则敏感性和特异性估计值可能存在偏倚。在本文中，我们推导出最大似然估计的敏感性和特异性时，一些患者的测试结果没有验证真实的疾病状态。
Sensitivity and specificity are two commonly used measures of accuracy for diagnostic tests. To calculate these two measures one has to know the true disease status of each patient. However, in practice sometimes some of the patients with test results are not selected for verification of disease status. Estimates of sensitivity and specificity may be biased if only patients who have verified disease statuses are used in the calculations. In this paper we derive Maximum Likelihood Estimators for sensitivity and specificity when some of the patients with test results do not have verified true disease statuses.