基于人工神经网络的老年男性血沉参考值与地理环境关系研究

Relationship Between Reference Value (Wintrobe) of Chinese Old Men's Erythrocyte Sedimentation Rate (ESR) and Geographical Factors Based on NN

以健康老年男性血沉参考值为例,收集中国各地用温氏法测定的健康老年男性血沉参考值,运用BP神经网络模拟分析其与地理环境的海拔高度、年日照时数、年平均相对湿度、年平均气温、年降水量等要素间的非线性、复杂性关系特征,通过5层神经网络,经过500次及1 300次自学习建立模拟规则后,用此规则很好地模拟健康老年男性血沉参考值与地理环境的关系。知道中国某地的地理要素,就可用此模型估算该地区老年男性血沉参考值。人工神经网络预测血沉参考值的模型自动定义具体的结构和参数,减少血沉参考值预测中的人为性,计算方便;同时,通过对不同地理要素的组合,模型可以寻找预测误差最低的地理要素组合关系。

This paper discusses the nonlinear relationship between reference value （wintrobe） standard of Chinese old men＇s ESR and geography environments based on artificial neural network. Reference value （wintrobe） standard of Chinese old men＇s ESR is connected with geographical environment, diet structure and settlement environment, etc. which interact with blood indirectly. It is artificial neural network that have the ability to obtain the nonlinear relationship between variables. In order to supply a basis for uniting the reference value （wintrebe） standard of Chinese old men＇s ESR, a research is made about the nonlinear relationship between the reference value （wintrobe） of Chinese healthy old men＇s ESR and five geographical factors, which are determined by the way of wintrobe laws based on artificial neural network. After building 5-layer NN, the relationship was simula- ted by artifical neural network（NN）. After training 500 times and 1300 times, the network works better. If the geographical factors known, the reference value （wintrobe） of Chinese old men＇s ESR can be simulated by the neural network more accurately.