The Impact Of Regional Geographic Environment On The Reference Value Of Right Heart Function

The right ventricle is an organ that plays an important role in the circulation of the lungs. It is involved in many diseases, for instance, it can cause the left ventricle dysfunction and affect the survival rate of patients with congenital heart disease before and after operation. In current studies on right ventricle, people mainly focus on its dependence on the physiological parameters of human body, such as the body surface area, pulmonary artery pressure, plasma brain natriuretic peptide concentra, thyroid function and so on. However, the studies on the effects of geographical factors on right ventricle are rarely reported. If the effects of geographical environment on the reference value of the right ventricle and the geographical distribution difference of right ventricle are ignored, it will affect the diagnostic accuracy.Three parameters of right ventricle:right ventricular diameter, right atrial diameter and right ventricle ejection fraction were filtered by age, sex, area, measuring method and instrument through literature search. According to the characteristics of the geographical environment in China, the topography, climate, soil factors were selected into the study, as follows:altitude （m, X₁）, longitude （°, X₂） and latitude (（, X₃）, the annual sunshine duration （h, X₄）, annual average temperature （℃, X₅）, annual average relative humidity （%, X₆）, the annual average wind speed （m/s, X₇）, soil organic matter content （% wt., X₈）, soil acid-base degree （X₉）, soil salinity （% wt, X₁0）. The relationship between the reference values of right ventricular parameters and geographical factors is quantitatively studied by using general correlation analysis and grey correlation analysis. By exploring the collinearity of independence variables, the linear models and non-linear models were built. Then, by testing the accuracy and the single factor variance of prediction models, the best fitting models were selected to predict the reference values of the right ventricular parameters in China. After the normal distribution test and the global trend analysis of the reference values of the right ventricular parameters in China were taken, the corresponding geostatistical methods were selected to make the spatial distribution map.The results of general correlation analysis show that altitude, latitude, annual average temperature and annual average relative humidity have significant effect on the reference values of right ventricular parameters. And the results of grey correlation show that the annual average temperature, annual average relative humidity, longitude, annual average wind speed and soil acid-base degree are the important factors effects the reference values of right ventricular parameters. Comprehensive two results, the longitude, latitude, altitude, annual average temperature, annual average relative humidity, annual average wind speed and soil acid-base degree were considered as the important factors of the reference values of right ventricular parameters. The results of model test show that the neural network model is the optimal prediction models of right ventricular parameters. The spatial distribution maps display a clear pattern that the reference values of right ventricular diameter and right atrial diameter is higher in the west than that in the east, and right ventricular ejection fraction is higher in the east than that in the west.Combined with qualitative and quantitative analysis, it is showed that the reference values of human right ventricular parameters is different in different geographical environment, indicated the relationship between right ventricular function and geographical environment can not be ignored. The partial least square method and the ridge regression model can solve the collinearity problem effectively. Moreover, nonlinear model is better reflects the nonlinear relationship between right ventricular parameters and geographical factors. The accuracy of nonlinear model is higher than the linear model, indicated there is a stronger nonlinear relationship between right ventricular parameters and geographical factors. The spatial distribution of right ventricular parameters has obvious change rules and shows a good consistency with the geographical environment. From the distribution maps, the reference values of right ventricular parameters can be read clearly, which can provide convenience for clinical workers.