血常规正常参考值与地理因素的非线性分析

发布时间：2018-10-10 17:10

【摘要】： 选题意义:血红蛋白和红细胞计数正常参考值指标在临床中有着至关重要的作用,可以用来确切的反应不同年龄不同性别的人是否有无贫血以及贫血的程度。而血红蛋白和红细胞计数与人们长期生活的地理环境(温度、湿度、海拔、日照、降水量、水质和土壤等)有着密切的关系,本论文研究了不同年龄不同性别的血红蛋白和红细胞计数正常参考值与地理因素的密切关系,揭示其分布规律,为临床的诊断提供可靠的科学依据。研究的理论依据、方法和过程:为探讨各血常规指标正常参考值与所选各种地理因素复杂的关系,本论文通过大量检索文献,以及向有关单位求购的方法,收集了全中国各地区不同年龄不同性别的血红蛋白和红细胞计数正常参考值的医学指标,主要为青年男性血红蛋白正常参考值、老年女性血红蛋白正常参考值、老年男性血红蛋白正常参考值、青年男性红细胞计数正常参考值、青年女性红细胞计数正常参考值、中年男性红细胞计数正常参考值、中年女性红细胞计数正常参考值和老年前期男性红细胞计数正常参考值8项指标。根据国家测绘局数据中心提供的共享资料,国家气象局数据中心提供的共享资料及有关地理著作、词典和相关文献,收集了五项地理因素指标:海拔高度(X_1)、年日照时数(X_2)、年平均气温(X_3)、年平均相对湿度(X_4)、年降水量(X_5)。对医学指标与这些地理学指标进行了线性和非线性的研究,从而建立它们之间的复杂线性模型和非线性BP神经网络模型。其线性模型主要为因子分析模型和趋势面分析模型,非线性模型为BP神经网络模型。应用以上的模型以及模型中的参数,代入了以选择1288个观测点的相应的地理因素指标,可计算出这1288个地方不同年龄不同性别的血红蛋白和红细胞计数正常参考值,借助GIS空间分析中的地统计分析模块,通过克里格(Kriging)插值法精确的内插出血常规指标正常参考值的地理分布图。研究取得的结论与创新之处:如果知道了中国某地的地理因素,就可用建立模型估算该地区不同年龄不同性别的血红蛋白和红细胞计数正常参考值,从地理分布图也可得到中国任何地方的正常参考值。在应用的三种模型里,发现应用非线性模型得出的精确度是最高的,更能很好地为临床诊断提供依据。本论文研究方法的创新之处在于对不同年龄段不同性别的人的血红蛋白和红细胞计数的正常参考值与地理因素进行了线性及非线性的定量化研究,与过去应用简单的相关分析和回归分析方法相比较,降低了人为的干扰,提高了模拟的精度。以中国整个区域为研究对象,借助GIS空间分析的方法,应用地统计分析模块中的克里格插值法,精确的绘制出正常参考值的趋势分布图。本论文的研究方法和研究成果将丰富临床医学,临床检验学,血液学,医学地理学,环境医学及生理学的内容。
[Abstract]:Significance: hemoglobin and normal reference values of red blood cell count play an important role in clinical practice, which can be used to determine whether or not people of different ages and sexes have anemia and the degree of anemia. Hemoglobin and red blood cell count are closely related to the geographical environment (temperature, humidity, altitude, sunshine, precipitation, water quality, soil, etc.) where people live for a long time. This paper studies the close relationship between the normal reference value of hemoglobin and red blood cell count of different ages and sexes and geographical factors, and reveals its distribution rule, which provides a reliable scientific basis for clinical diagnosis. The theoretical basis, method and process of the study: in order to explore the complex relationship between the normal reference value of each blood routine index and various geographical factors selected, this paper through a large number of retrieval literature, as well as the method of seeking purchase from relevant units, The normal reference values of hemoglobin and red blood cell count of different ages and sexes in all regions of China were collected, mainly for young men and elderly women. The normal reference value of hemoglobin in old men, the normal reference value of red blood cell count in young men, the normal reference value of red blood cell count in young women, the normal reference value of red blood cell count in middle-aged men, The normal reference value of red blood cell count in middle aged women and the normal reference value of red blood cell count in male in pre-old age were 8 indexes. According to the shared information provided by the data center of the State Survey and Mapping Bureau, the shared data provided by the data center of the National Meteorological Administration and relevant geographical works, dictionaries and related documents, Five geographical factors were collected: altitude (X _ 1), annual sunshine hours (X _ S _ 2), annual mean temperature (X _ S _ 3), annual average relative humidity (X _ 4), annual precipitation (X _ 5). In this paper, the linear and nonlinear models of medical indexes and these geographical indexes are studied, and the complex linear models and nonlinear BP neural network models between them are established. The linear model is mainly factor analysis model and trend surface analysis model, and the nonlinear model is BP neural network model. Using the above model and the parameters in the model, the corresponding geographical factor indexes of 1288 observation points were added to calculate the normal reference values of hemoglobin and red blood cell count for different ages and genders in these 128 places. With the help of geostatistical analysis module in GIS spatial analysis, the geographical distribution map of normal reference value of blood routine index is accurately interpolated by Kriging (Kriging) interpolation method. Conclusions and innovations of the study: if we know the geographical factors in a certain place in China, we can establish a model to estimate the normal reference values of hemoglobin and red blood cell count for different ages and genders in this area. Normal reference values for any part of China can also be obtained from geographical maps. Among the three models applied, it is found that the accuracy of the nonlinear model is the highest, which can provide the basis for clinical diagnosis. The innovation of this research method lies in the linear and nonlinear quantitative study of the normal reference values and geographical factors of hemoglobin and red blood cell count of people of different ages and different genders. Compared with the simple correlation analysis and regression analysis, the artificial interference is reduced and the simulation accuracy is improved. Taking the whole region of China as the research object, using the method of GIS spatial analysis and using the Kriging interpolation method in the geostatistical analysis module, the trend distribution map of the normal reference value is drawn accurately. The research methods and results of this paper will enrich the contents of clinical medicine, clinical laboratory, hematology, medical geography, environmental medicine and physiology.
【学位授予单位】：陕西师范大学
【学位级别】：硕士
【学位授予年份】：2008
【分类号】：R188

【参考文献】