Research of the Cluster and Discriminant Analysis in the Myocardial Infarction Electrocardiogram
Objective : To study the clinical indexes of the myocardial infarction(MI) by the ECG, and find more effective techniques of the clinical diagnosis．Method: 94 cases of the ECG records are selected，where 38 of which is acute myocardial infarction group(AMI), 11 of subacute myocardial infarction group (SMI), 19 obsolete myocardial infarction group(OMI), and 26 of which is control group. The following 7 indexes are collected from the 94 records respectively: age, gender, chest pain, diaphorase measurement, abnormal Q wave, elevational ST segment, inverse T wave etc. Firstly, we find the classification of the above indexes according to the hierarchical clustering analysis, using the world famous statistics software SPSS. Then give the discriminant function and the effect analysis according to the stepwise discriminant analysis. Result: (1) The computation shows that, all factors are classified into two categories: one is constituted of the age, elevational ST segment, inverse T wave; another one is constituted of the rest 4 indexes; (2) According to the stepwise discriminant analysis, the following 4 indexes are screened out as combined indexes of the discrimination function: chest pain, diaphorase measurement, abnormal Q wave and elevational ST segment, which coincides with the results of the clustering analysis. Finally the check computation shows that, the overall discrimination sensitivity is 64.9%, and the discriminant sensitivities of the AMI, SMI, OMI, control are 57.9%、 54.5%、47.4%. 、92.3%. Conclusion: The combined indexes constituted by chest pain, diaphorase measurement, abnormal Q wave and elevational ST segment, play an important role in the diagnosis of the MI．
The Myocardial Infarction Diaphorase Measurement The Clustering Analysis The Stepwise Discriminant Analysis
LIU Guangchen LI Zhaohui HAN Chunlei SONG Mei
School of Mathematics&Informatics, Lu Dong University, Yantai, P.R.China, 264025 Hospital of Lu Dong University, Yantai P.R.China, 264025 School of Health and Management, Binzhou Medical University, Yantai, P.R.China, 264025
国际会议
2008年国际应用统计学术研讨会(2008 International Institute of Applied Statistics Studies)
烟台
英文
1-5
2008-08-14（万方平台首次上网日期，不代表论文的发表时间）