摘要
在现有公共卫生体系基础上,提出了建立公共卫生事件监测与预警系统的框架模型,并首次利用神经网络的基本原理,将改进的BackPropagation算法应用于系统的核心预测模型。该系统可以通过监测医疗数据的变化情况来迅速预测出疾病的发生和未来的发展趋势,经初步模拟研究,预测精度可达93%,为公共卫生事件的长期可预测性提供了一种新的途径。
A predictive warning system framework of public health accidents is proposed based on the current public health system. Improved Back-Propagation algorithm is applied to the predictive model according to the elementary principle of artificial neural network for the first time. Using this system, the outbreak and tendency of some epidemics can be forecasted through the change of medical data rapidly. From the preliminary simulations, it could be observed that the accuracy of prediction can reach 93 percent, which shows it provides a new way for long-term forecasting of public health.
出处
《计算机应用研究》
CSCD
北大核心
2005年第6期165-167,178,共4页
Application Research of Computers
基金
武汉大学软件工程国家重点实验室开发基金资助项目