基于改进BP神经网络模型的肺结核发病率预测

发布时间：2018-11-23 11:59

【摘要】：目的建立用于肺结核发病率预测的人工神经网络模型,预测肺结核疫情发生发展趋势,为肺结核的预防和控制提供理论依据。方法选取肺结核2000~2014年发病率数据,采用改进的误差反向传播(BP)神经网络算法建立预测模型。其中以2000~2013年的发病率数据作为训练样本,以2014年的发病率数据来检验模型的有效性。并对2015~2019年肺结核的发病率进行预测。结果采用单隐层神经网络模型,输入层节点数为3,隐含层节点数为7,输出层节点数为1。建立的肺结核发病率预测模型在仿真预测样本处的平均相对误差为0.7597%,在检验样本处的相对误差为0.2649%。经预测,2015~2019年肺结核的发病率分别为69.33/10万、71.16/10万、64.49/10万、62.41/10万和72.78/10万。结论采用改进的BP神经网络算法建立的肺结核发病预测模型具有较高的预测精度及较低的预测相对误差,为肺结核疫情预测提供一种新的预测模型。
[Abstract]:Objective to establish an artificial neural network (Ann) model for predicting the incidence of pulmonary tuberculosis (TB), and to predict the trend of the occurrence and development of TB, and to provide theoretical basis for the prevention and control of pulmonary tuberculosis. Methods the incidence data of pulmonary tuberculosis from 2000 to 2014 were selected and the prediction model was established by using the improved error back-propagation (BP) neural network algorithm. The incidence data from 2000 to 2013 were used as training samples, and the incidence data of 2014 were used to test the validity of the model. The incidence of pulmonary tuberculosis from 2015 to 2019 was predicted. Results using the single hidden layer neural network model, the number of nodes in the input layer is 3, the number of hidden layer nodes is 7, and the number of output layer nodes is 1. The average relative error of the model for predicting the incidence of pulmonary tuberculosis is 0.7597 at the simulated prediction sample and 0.2649 at the test sample. The incidence of pulmonary tuberculosis in 2015-2019 is predicted to be 69.33 / 100,000, 711.6 / 100,644.49 / 100,624.1 / 100,000 and 727.8 / 100, respectively. Conclusion the prediction model of pulmonary tuberculosis based on improved BP neural network has higher prediction accuracy and lower relative error, which provides a new prediction model for the prediction of tuberculosis epidemic situation.
【作者单位】：河南中医药大学基础医学院;郑州大学第一附属医院血液科;
【基金】：河南省软科学研究重点项目(No:102400440002) 河南中医学院科研苗圃工程项目(No:MP2014-07)
【分类号】：R521

【相似文献】