摘要
目的探讨贝叶斯判别分析在诊断原发性醛固酮增多症中的价值。方法收集2008年7月至2013年12月近5年重庆医科大学附属第一医院疑诊原发性醛固酮增多症(原醛)病例321例,分为原发性醛固酮增多症组和原发性高血压组。利用贝叶斯判别函数获取原醛分类函数系数及原醛后验概率,根据受试者工作特征曲线(ROC),获取最佳切点。进一步对此法及立位血浆醛固酮/肾素浓度比值(ARR)在原醛筛查中的价值进行对比。结果原醛后验概率为0.34时,其诊断原醛的灵敏度和特异度最高,分别为0.84、0.84,立位ARR切点取43.45时灵敏度和特异度达最高,分别为0.88、0.76,原醛后验概率的ROC曲线下面积为0.911(0.870-0.951),大于立位ARR曲线下面积0.893(0.845-0.941)。结论贝叶斯判别分析诊断原醛的价值优于立位ARR。
Objectives To explore the value of Bayesian discriminant analysis in clinical diagnosis of primary aldosteronism（PA）. Methods Three hundred and twenty-one suspected cases of PA were enrolled and divided into two groups： PA group and essential hypertension group. Bayesian probability discriminant analysis was used to get the original classification function coefficients and posterior probability of PA. Then, best cut-off point was obtained according to the receiver operating characteristic （ROC） curves. Finally the value of Bayesian discriminant analysis in diagnosis of PA were compared with that of upright plasma aldosterone/renin ratio （ARR）. Results When the posterior probability is 0.34, the accuracy of Bayesian analysis in diagnosis of PA reach to the highest level, sensitivity and specificity are 0.84, 0.84 respectively. Upright ARR take 43.45 as cut-offpoint reaches the highest sensitivity and specifici~ respectively 0.88, 0.76.Area under the curve of the posterior probability of PA based on Bayesian discriminant equation are 0.911 （0.870-0.951）, higher than that of upright ARRs0.893（0.845-0.941）. Conclusion The value of Bayesian discriminant function in the diagnosis of PA is better than upright ARK.
出处
《中国实用内科杂志》
CAS
CSCD
北大核心
2015年第10期835-838,共4页
Chinese Journal of Practical Internal Medicine
基金
国家临床重点专科建设项目[(2011)170号]
关键词
贝叶斯判别分析
原发性醛固酮增多症
诊断
血浆醛固酮/肾素浓度比值
Bayes discriminant analysis
primary aldosteronism
diagnosis
plasma aldosterone concentration/renin concentration ratio
