基于因子BP神经网络预测模型的胃溃疡发病率预测

目的:在医学卫生领域,疾病发病率受许多因素的影响,很难用结构式因果模型解释,根据神经网络预测是一种行之有效的方法。多层前馈神经网络(BP)应用于疾病发病率的预测时,由于影响它的气象因素,如月平均气温、月平均气压、月平均相对湿度等本身具有很大相关性,且维数较高,BP神经网络的预测精度会下降。方法:针对这一问题,提出了利用因子分析方法对原输入空间进行重构,并根据各主成分的贡献率确定网络结构,有效地解决了预测精度下降的问题。结果:以2003年1月至2009年12月青海海西州地区胃溃疡发病率的资料验证了该方法的有效性。结论:应该充分考虑胃溃疡在各时间段的发病率特征,以便更有重点地进行健康防治工作,有效地降低胃溃疡对人类的危害,保障人类的生活品质。

Gastric Ulcer Incidence Rate Forecasting Based on Factor BP Neural, Network Prediction Model

In medicine,due to the fact that diseases are often affected by many factors,it is difficult to use a structural causal model.On the other hand,it would be effective to establish a dynamic model,based on their time-series changes.To predict the incidence rate of diseases,because of meteorological factors,such as the monthly mean temperature,monthly average atmospheric pressure,and monthly mean relative humidity are strongly correlated between themselves and with very high dimensions,the accuracy of the Radial back propagation(BP) neural network prediction may be very low.Methods: To solve this problem,the use of the Principal Component Analysis method to reconstruct the original input space is proposed,and,based on the contribution rate of all main components to determine the network structure,the problem of low prediction accuracy is effectively solved.Results: The incidence rate of gastric ulcer,from January 2003 to December 2009,in Qinghai Haixizhou region are used to validate the method.Conclusion: The gastric ulcer incidence rate characteristics in each time period should be duly considered in order to carry out a more focused health prevention and treatment and to effectively reduce the hazards of gastric ulcer.