冠心病患者血清中微量元素的原子吸收法测定及人工神经网络分析  

The Determination of Trace Elements in Coronary Disease Patient Serum by AAS Method and Analysis by Backpropagation Algorithm for Neural Metwork

作　　者：蒋淑梅[1,2] 柯以侃[1,2] 时彦 迟锡曾 周宏珠[3] 

机构地区：[1]北京化工大学应化系 [2]北京师范大学化学系 [3]北京积水潭医院

出　　处：《分析试验室》1999年第2期42-43,共2页Chinese Journal of Analysis Laboratory

摘　　要：应用火焰原子吸收法和石墨炉原子吸收法测定了冠心病患者和对照组（健康人）血清中Ｚｎ、Ｃｕ、Ｆｅ、Ｃａ、Ｍｇ、Ｃｒ、Ｍｎ和Ｓｒ等的含量。对所得数据用反向传播神经网络（Ｂ－Ｐ法）进行分析，建立了冠心病的神经网络识别系统，预报识别率达１００％，可作为该病诊断的一种有效的辅助手段。The methods of FAAS and GFAAS were applied to determine the contents of Zn,Cu,Fe, Ca,Mg,Cr,Mn and Sr in serum of coronary disease patients and healthy people.The data from these methods were analysed by backpropagation algorithm for neural network and an identification system of was established .The recoveries are 100% for the prediction set of samples and it can be an efficient aid to diagnose this kind of disease.

关 键 词：铬 铜 镁 锌 冠心病 血清 微量元素 模式识别 

分 类 号：R446.112[医药卫生—诊断学] R541.404[医药卫生—临床医学]