血压指标的聚类分析

The cluster analysis of variables of blood pressure

利用353人的血压数据,运用2种不同的聚类分析方法对15个血压指标分类并挑选典型指标:一是利用SAS软件包中提供的变量聚类过程(VARCLUS过程)对指标进行裂类聚类分析;一是利用SAS中提供的样品聚类过程(CLUSTER过程)对指标进行系统聚类分析.2种方法得到了相同的结果:15个指标分成了4类,得到4个典型指标:BPR(收缩压与舒张压比值),BPP(收缩压与舒张压乘积),MAPPr(平均动脉压与脉率乘积),BPRPr(BPR与脉率乘积).

Analysed fifteen variables of blood pressure and got the typical variables by using two different cluster analysis methods to three hundred and fifty-three people's data of blood pressure . One method used the VARCLUS process in SAS(Statistical Analysis System), and the other method used the CLUSTER process in SAS . The two different methods achieved the same result.The fifteen variables were divided into four groups , and got four typical variables : BPR(the ratio of the systolic pressure to the diastolic pressure), BPP(the product of the systolic pressure and the diastolic pressure), MAPPR(the product of the mean arterial pressure and the speed of pulse), BPRPR(the product of BPR and the speed of pulse).