Background Most indices for evaluating a diagnostic test can be expressed as functions of sensitivity （SEN） and specificity （SPE）. Practically, all existing methods suffer from the inability to weight sensitivity ...Background Most indices for evaluating a diagnostic test can be expressed as functions of sensitivity （SEN） and specificity （SPE）. Practically, all existing methods suffer from the inability to weight sensitivity and specificity relative to their importance. In this paper, we developed a novel index, the weighted Youden index, that allows Youden index to be a combination of sensitivity and specificity with user-defined weights. Methods The weighted Youden index Jw is defined as Jw=2（wxSEN＋（1-w）SPE）-I （0 〈w 〈1）. It has three properties： （1） the sum of the weights which are attached to sensitivity and specificity should be equal to 1; （2） the range of Jw should be within [-1, 1], which is the range of the Youden index J; （3） Jw should be equal to J when sensitivity and specificity have equal weights. According to the central limit theorem, we obtain the standard error of Jw, and propose a statistical inference method to compare two weighted Youden indices. The monotonicity of the test statistic was discussed. Results An example of comparing two diagnostic tests for pheochromocytoma was used to demonstrate the weighted Youden index method. Weighted Youden index, the confidence interval for each test and the hypothesis test of comparing two independent diagnostic tests were presented. Assigning the weights is essential to the weighted Youden index approach. Conclusion The weighted Youden index can broaden further research in weighting sensitivity and specificity exp ts applications in diagnostic test development and motivate icitly.展开更多
It is not easy to construct a model to describe the geochemical background in geochemical anomaly detection due to the complexity of the geological setting.Isolation forest and its improved algorithms can detect geoch...It is not easy to construct a model to describe the geochemical background in geochemical anomaly detection due to the complexity of the geological setting.Isolation forest and its improved algorithms can detect geochemical anomalies without modeling the complex geochemical background.These methods can effec-tively extract multivariate anomalies from large volume of high-dimensional geochemical data with unknown population distribution.To test the performance of these algorithms in the detection of mineralization-related geochemical anomalies,the isolation forest,extended isolation forest and generalized isolation forest models were established to detect multivariate anomalies from the stream sediment survey data collected in the Wu-laga area in Heilongjiang Province.The geochemical anomalies detected by the generalized isolation forest model account for 40%of the study area,and contain 100%of the known gold deposits.The geochemical anomalies detected by the isolation forest model account for 20%of the study area,and contain 71%of the known gold deposits.The geochemical anomalies detected by the extended isolation forest algorithm account for 34%of the study area,and contain 100%of the known gold deposits.Therefore,the isolation forest mo-del,extended isolation fo-rest model and generalized isolation forest model are comparable in geochemical anomaly detection.展开更多
目的·采用约登指数改进肝纤维化诊断模型的性能,解决组间样本数相差大时诊断灵敏度不平衡的问题。方法·使用在GitHub网站公开获取的来自上海中医药大学附属曙光医院的482例乙型肝炎病毒(hepatitis B virus,HBV)感染患者和来...目的·采用约登指数改进肝纤维化诊断模型的性能,解决组间样本数相差大时诊断灵敏度不平衡的问题。方法·使用在GitHub网站公开获取的来自上海中医药大学附属曙光医院的482例乙型肝炎病毒(hepatitis B virus,HBV)感染患者和来自厦门市中医院的86例HBV感染患者分别作为训练集和验证集开展研究。基于HBV患者的年龄和3项血液学检查结果(血小板计数、血清中谷草转氨酶和谷丙转氨酶含量),建立线性判别分析、随机森林、梯度增强、决策树4种机器学习模型,实现早期和晚期肝纤维化的诊断,以及肝纤维化和肝硬化的诊断。借助约登指数调整模型的分类阈值和诊断结果。采用总准确率、受试者工作特征曲线下面积(area under the curve,AUC)和灵敏度等指标,比较各模型以及临床常用的基于4个因素的纤维化指数(fibrosis index based on the 4 factor,FIB-4)的诊断性能。结果·在肝纤维化诊断中,4种机器学习模型均存在组间灵敏度不平衡的现象。在向模型引入约登指数后,组间灵敏度的差别均大幅减小;机器学习模型的总准确率和AUC普遍高于FIB-4。结论·基于约登指数的诊断模型可平衡各组间的灵敏度,有助于提高肝纤维化诊断模型的综合性能。展开更多
基金This work was supported by the grants from the National Natural Science Foundation of China （No. 30972554） and the Natural Science Foundation of Guangdong Province （No. 9 ! 5180200400001）.
文摘Background Most indices for evaluating a diagnostic test can be expressed as functions of sensitivity （SEN） and specificity （SPE）. Practically, all existing methods suffer from the inability to weight sensitivity and specificity relative to their importance. In this paper, we developed a novel index, the weighted Youden index, that allows Youden index to be a combination of sensitivity and specificity with user-defined weights. Methods The weighted Youden index Jw is defined as Jw=2（wxSEN＋（1-w）SPE）-I （0 〈w 〈1）. It has three properties： （1） the sum of the weights which are attached to sensitivity and specificity should be equal to 1; （2） the range of Jw should be within [-1, 1], which is the range of the Youden index J; （3） Jw should be equal to J when sensitivity and specificity have equal weights. According to the central limit theorem, we obtain the standard error of Jw, and propose a statistical inference method to compare two weighted Youden indices. The monotonicity of the test statistic was discussed. Results An example of comparing two diagnostic tests for pheochromocytoma was used to demonstrate the weighted Youden index method. Weighted Youden index, the confidence interval for each test and the hypothesis test of comparing two independent diagnostic tests were presented. Assigning the weights is essential to the weighted Youden index approach. Conclusion The weighted Youden index can broaden further research in weighting sensitivity and specificity exp ts applications in diagnostic test development and motivate icitly.
文摘It is not easy to construct a model to describe the geochemical background in geochemical anomaly detection due to the complexity of the geological setting.Isolation forest and its improved algorithms can detect geochemical anomalies without modeling the complex geochemical background.These methods can effec-tively extract multivariate anomalies from large volume of high-dimensional geochemical data with unknown population distribution.To test the performance of these algorithms in the detection of mineralization-related geochemical anomalies,the isolation forest,extended isolation forest and generalized isolation forest models were established to detect multivariate anomalies from the stream sediment survey data collected in the Wu-laga area in Heilongjiang Province.The geochemical anomalies detected by the generalized isolation forest model account for 40%of the study area,and contain 100%of the known gold deposits.The geochemical anomalies detected by the isolation forest model account for 20%of the study area,and contain 71%of the known gold deposits.The geochemical anomalies detected by the extended isolation forest algorithm account for 34%of the study area,and contain 100%of the known gold deposits.Therefore,the isolation forest mo-del,extended isolation fo-rest model and generalized isolation forest model are comparable in geochemical anomaly detection.
文摘目的·采用约登指数改进肝纤维化诊断模型的性能,解决组间样本数相差大时诊断灵敏度不平衡的问题。方法·使用在GitHub网站公开获取的来自上海中医药大学附属曙光医院的482例乙型肝炎病毒(hepatitis B virus,HBV)感染患者和来自厦门市中医院的86例HBV感染患者分别作为训练集和验证集开展研究。基于HBV患者的年龄和3项血液学检查结果(血小板计数、血清中谷草转氨酶和谷丙转氨酶含量),建立线性判别分析、随机森林、梯度增强、决策树4种机器学习模型,实现早期和晚期肝纤维化的诊断,以及肝纤维化和肝硬化的诊断。借助约登指数调整模型的分类阈值和诊断结果。采用总准确率、受试者工作特征曲线下面积(area under the curve,AUC)和灵敏度等指标,比较各模型以及临床常用的基于4个因素的纤维化指数(fibrosis index based on the 4 factor,FIB-4)的诊断性能。结果·在肝纤维化诊断中,4种机器学习模型均存在组间灵敏度不平衡的现象。在向模型引入约登指数后,组间灵敏度的差别均大幅减小;机器学习模型的总准确率和AUC普遍高于FIB-4。结论·基于约登指数的诊断模型可平衡各组间的灵敏度,有助于提高肝纤维化诊断模型的综合性能。
