基于主成分分析的左室Tei指数参考值的地理分布

摘要

Objective: To provide a scientific standard of left ventricular Tei index for healthy people from various region of China, and to lay a reliable foundation for the evaluation of left ventricular diastolic and systolic function. Methods: The correlation and principal component analysis were used to explore the left ventricular Tei index, which based on the data of 3 562 samples from 50 regions of China by means of literature retrieval. hTe nine geographical factors were longitude(X1), latitude(X2), altitude(X3), annual sunshine hours (X4), the annual average temperature (X5), annual average relative humidity (X6), annual precipitation (X7), annual temperature range (X8) and annual average wind speed (X9). ArcGIS sotfware was applied to calculate the spatial distribution regularities of letf ventricular Tei index. Results: hTere is a signiifcant correlation between the healthy people’s letf ventricular Tei index and geographical factors, and the correlationcoeffcients were 0.107 (r1), 0.301 (r2), 0.029 (r3), 0.277 (r4),−0.256(r5),−0.289(r6),−0.320(r7), 0.310 (r8) and 0.117 (r9), respectively. A linear equation between the Tei index and the geographical factor was obtained by regression analysis based on the three extracting principal components. hTe geographical distribution tendency chart for healthy people’s letf Tei index was iftted out by the ArcGIS spatial interpolation analysis. Conclusion: hTe geographical distribution for letf ventricular Tei index in China follows certain pattern. hTe reference value in North is higher than that in South, while the value in East is higher than that in West.%目的：为中国不同地区的健康人左心室Tei指数参考值提供一个科学依据，为左室舒张和收缩的整体功能的评价提供一个较为可靠的判断依据。方法：通过文献检索收集来自50个市(县)的3562例健康人左室Tei指数，将其与9项地理指标，即经度(X1)、纬度(X2)、海拔(X3)、年日照时数(X4)、年平均气温(X5)、年平均相对湿度(X6)、年降水量(X7)、气温年较差(X8)、年平均风速(X9)，进行简单相关分析和主成分分析；应用ArcGIS10.0软件探索左室Tei指数的空间分布规律。结果：健康人左室Tei指数与9项地理指标之间存在显著相关性，相关系数分别为r1=0.107， r2=0.301，r3=0.029，r4=0.277，r5=−0.256，r6=−0.289，r7=−0.320，r8=0.310，r9=0.117。根据提取的3个主成分与Tei指数进行回归分析，获得了Tei指数与地理指标间的线性模型。应用ArcGIS10.0软件进行插值分析，拟合出健康人左室Tei指数的空间分布趋势图。结论：中国人左室Tei指数参考值存在一定的地理分布规律——西高东低，北高南低。