冠心病患者运动试验Bayes准则模型的应用

Application of Bayes Discriminant Model in Exercise Testing for Patients with Coronary Artery Disease

本文采用无创性Bayes准则多元逐步判别分析方法,对73例同期接受改良Bruce方案平板运动试验及选择性冠状动脉造影对照的住院病人,进行回顾性资料处理(SAS/PC 6.03版微机软件系统),并将所建立的拟合模型对29例患者进行前瞻性研究.结果显示该多变量分析方法在建立良好的预测模型和判定指标的筛选方面具有独特价值,提示多变量分析方法可显著提高运动试验诊断冠心病的敏感性和特异性.

Multivariate analysis of Bayes' Theo-rem was used for the diagnosis of coronary artery dis-ease. AH 73 cases in the corresponding period under-went limited exercise testing following modified Bruce protocol 2 weeks prior to selective coronary an-giography which was adopted as the diagnostic stan-dard. The software package SAS/PC 6. 03 was used to establish the diagnostic model. 29 cases were then studied prospectively. The results showed that this multivariate analysis has good and special value in construction of predictive model and the selection of discriminant indices, and also could highly improve the diagnostic ability of exercise testing.