中国人肺活量正常参考值与地理因素的关系

发布时间：2018-04-09 20:43

本文选题：肺活量　切入点：参考值　出处：《陕西师范大学》2009年硕士论文

【摘要】： 肺活量正常参考值是肺功能检查的一项重要指标,可以用来判断限制性通气功能障碍程度。在临床上,对肺活量正常参考值的增加或是减少的解释多以病理学为主,生理学解释也多以身高、体重、年龄以及性别等为主,很少涉及人类生存环境地形、气候、地貌等因素对肺部器官功能的影响。而本论文的创新之处在于从地理环境方面出发,对不同性别各年龄段的中国人肺活量正常参考值与地理因素,进行了线性与非线性的定量化研究与对比。经过充分的检索发现,虽有多篇论文报道了肺活量正常参考值与海拔、温度、湿度、风速等地理因素的关系,但还只是定性的描述或用线性回归分析的方法进行简单的线性分析,像本论文这样对肺活量正常参考值与地理因素进行专题化、定量化、线性与非线性化的研究,从整个地理环境着手,系统的以中国这一大的区域来研究,还未见报道。为探讨不同性别各年龄段肺活量正常参考值与所选七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肺活量正常参考值。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,收集了七项地理因素指标: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)、气温年较差(X_6)、年平均风速(X_7)。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多元回归模型、偏最小二乘回归模型、岭回归模型、非线性回归模型、BP人工神经网络模型。最终通过模型对比研究,寻找出肺活量正常参考值的最优预测模型: ①预测男性儿童肺活量正常参考值运用人工神经网络模型,构建5层神经网络,训练次数为264次时,模型最优; ②预测女性儿童肺活量正常参考值运用偏最小二乘回归模型,其模型为:Y=1321.21-0.043514X_1+0.010103X_2-0.25246X_3-0.75417X_4-0.013527X_5+3.83186X_6+38.61289X_7±133.50; ③预测青春期男性肺活量正常参考值运用偏最小二乘回归模型,其模型为:(?)=3024.77-0.131640X_1+0.022388X_2+2.381548X_3-1.344117X_4-0.030850X_5+7.881706X_6+39.820870X_7±557.93; ④预测青春期女性肺活量正常参考值运用人工神经网络模型,构建5层神经网络,训练次数为3000次时,模型最优; ⑤预测青年男性肺活量正常参考值运用非线性回归模型,其模型为:(?)=4890.41-0.03057X_1-0.094425X_2+27.2634X_3-1.3680X_3~2-12.4397X_4-0.1151X_5-3381.2805/X_6+18.1272X_7±255.69; ⑥预测青年女性肺活量正常参考值运用偏最小二乘回归模型,其模型为:(?)=2692.70-0.001604X_1+0.017849X_2-2.28365X_3-1.42546X_4-0.029721X_5+1.94182X_6+12.64521X_7±114.90; ⑦预测中年男性肺活量正常参考值运用非线性回归模型,其模型为:(?)=-1702.56+0.2226X_1+0.1971X_2-15.6680X_3+48.4351X_4-0.4567X_5+153.8167X_6-3.1040X_6~2+109.3521X_7±350.03; ⑧预测中老年女性肺活量正常参考值运用多元线性回归模型,其模型为:(?)=3105.831-20.085X_6±230.427; ⑨预测老年男性肺活量正常参考值运用非线性回归模型,其模型为:(?)=3184.086+171213.7967/X_5±300.33。应用以上的模型,代入了已选择4383个观测点的相应的地理因素指标,可计算出这4383个地方不同年龄不同性别的肺活量正常参考值,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肺活量正常参考值的空间趋势分布图。如果知道了中国某地的地理因素,就可用此模型估算该地区不同年龄不同性别的肺活量正常参考值,从地理分布图也可得到中国任何地方的正常参考值。
[Abstract]:The normal reference value of lung activity is an important index of pulmonary function examination , which can be used to judge the degree of limiting ventilation dysfunction .

In order to study the relationship between normal reference value and seven geographical factors of different age groups in different age groups , the author collected seven geographical factors , such as altitude ( X _ 1 ) , annual sunshine duration ( X _ 2 ) , annual average temperature ( X _ 3 ) , annual average relative humidity ( X _ 4 ) , annual precipitation ( X _ 5 ) , annual average temperature ( X _ 6 ) and annual average wind speed ( X _ 7 ) .

( 1 ) Using artificial neural network model to predict the normal reference value of pulmonary activity of male children , construct 5 - layer neural network , the training times are 264 times , the model is optimal ;

( 2 ) The regression model was used to predict the normal reference value of pulmonary activity in female children . The model was Y = 130.21 - 0.043514X _ 1 + 0.010103X _ 2 - 0.25246X _ 3 - 0.75417X _ 4 - 0.013527X _ 5 + 3.83186X _ 6 + 38.61289X _ 7 卤 133.50 ;

( 3 ) A partial least squares regression model was used to predict the normal reference value of pulmonary activity in adolescents . The model was : ( ? ) = 3024.77 - 0.131640X _ 1 + 0.022388X _ 2 + 2.381548X _ 3 - 1.344117X _ 4 - 0.030850 X _ 5 + 7.881706X _ 6 + 39.820870X _ 7 卤 557.93 ;

( 4 ) The artificial neural network model was applied to predict the normal reference value of pulmonary activity of adolescent female , and five - layer neural network was constructed , and the training frequency was 3000 times , and the model was optimal ;

( 5 ) A nonlinear regression model is used to predict the normal reference value of pulmonary activity in young men . The model is : ( ? ) = 4890.41 - 0.03057X _ 1 - 0.094425X _ 2 + 27.2634X _ 3 - 1.3680X _ 3 ~ 2 - 12.4397X _ 4 - 0.ANG X _ 5 - 3381.2805 / X _ 6 + 18.1272X _ 7 卤 255.69 ;

( 6 ) The model is : ( ? ) = 2692.70 - 0.001604X _ 1 + 0.017849X _ 2 - 2.28365X _ 3 - 1.42546X _ 4 - 0.029721X _ 5 + 1.94182X _ 6 + 12.64521X _ 7 卤 14.90 ;

( 7 ) A nonlinear regression model was used to predict the normal reference value of pulmonary activity in middle - aged men . The model was : ( ? ) = - 1702 . 56 + 0 . 2226X _ 1 + 0.1971X _ 2 - 15.6680X _ 3 + 48.435 1X _ 4 - 0 . 4567X _ 5 + 153.8167X _ 6 - 3.1040X _ 6 ~ 2 + 109.3521X _ 7 卤 350.03 ;

( 8 ) A multivariate linear regression model was used to predict the normal reference value of pulmonary activity in middle - aged and old women . The model was ( ? ) = 310.831 - 20.085X _ 6 卤 230.427 ;

9 . A nonlinear regression model was used to predict the normal reference value of pulmonary activity in elderly men . The model was : ( ? ) = 3184 . 086 + 171213 . 7967 / X _ 5 卤 300 . 33 .

By means of the above model , the corresponding geographical factor indexes of 4383 observation points have been selected , and the normal reference value of pulmonary activity with different ages in these 4383 places can be calculated . By means of the geostatistical analysis module in GIS spatial analysis , the spatial trend distribution map of the normal reference value of pulmonary activity is accurately interpolated by Kriging interpolation method . If the geographical factors of a certain place in China are known , the normal reference value of different ages of different ages in the region can be estimated by this model , and the normal reference value of any place in China can be obtained from the geographical distribution map .

【学位授予单位】：陕西师范大学
【学位级别】：硕士
【学位授予年份】：2009
【分类号】：R188

【引证文献】