摘要
为探讨高频心电图对冠心病的判别作用,采用Bayes方法分析93例冠心病患者和77例正常人体表12导联的高频切迹。结果显示:除Ⅰ和V_2导联外,其余10个导联的高频切迹对冠心病和正常人具有明显的判别作用(P<O.01)。由被引入的10个导联高频切迹数建立的两个判别函数对两组人群具有非常明显的判别能力(P<0.01)。判别效果评价表明,无论是内部检验,还是外推检验,其判别符合率、敏感性和特异性均较高,分别达88.82％、89.25％、88.31％和84.11％、83.87％、84.42％。认为高频切迹对冠心病具有较好的判别作用,对冠心病的早期诊断具有一定的指导意义。
Adopting Bayes discriminant analysis, authors studied high frequency notching (HFN) of body surface, 12 leads in 93 cases of coronary heart disease (CHD)and 77 healthy subjects in order to in-vestigate discriminant effect of HFN on CHD. Results showed that HFN of 10 leads had significant discriminant effects except for I and V2 leads among 12 leads by stepwise discriminant analysis (P<0.01) . Meanwhile,we established two discriminant functions of CHD and healthy subjects by meams of 10 leads , introducedwith sig-nificantlydiscriminant ability for CHD and normal group(P < 0.01)Coincidence rate, sensitivity and specificity of either internal or external determination came high up to 88.82%, 89.25%, 88.31%and 84.11%, 83.87%, 84.42%, respectively.These results suggest that HFECG possessed significant discriminant ability for CHD,and it could play an important role in early diagnose of CHD.
出处
《心电学杂志》
1998年第4期202-203,212,共3页
Journal of Electrocardiology(China)
关键词
高频心电图
高频切迹
冠心病
判别分析
High frequency electrocardiogram High fre-quency notching Coronary heart disease Discriminant analysis
