Abstract：Objective Hopfield neural network model was established to assessment the risk of sporadic measles outbreaks and identify high risk areas. Methods 9 indicators were determined as parameters to model the Hopfield neural network using Matrix lab software. The indicators included four types of the incidence, vaccination rate, running quality of measles monitoring system, and disposition of epidemic spot. Results The level of the risk of sporadic measles outbreaks in Xiangzhou District was identified as "extreme high risk". The risk level of Jinwan District, Doumen District was "high risk"and"low risk",respectively. Conclusions Hopfield neural network model could be used to assessment the risk of sporadic measles outbreaks and identify high risk areas preliminarily.
[ 1] 迮文远.计划免疫学[M].第2版.上海:上海科技出版社,2001: 359. [2 ] 许振国,陈双艳,高子厚,等.传染病风险评估体系研究进展[J].疾病监测,2012,27（10）:834-837. [3 ] Hopfield JJ. Neural networks and physical systems with emergent collective computational abilities.[J]. Proc Natl Acad Sci U S A.1982,79(8): 2554–2558. [4 ] 卫生部.关于做好2011年消除麻疹工作的通知[Z].2011-04-18. [ 5] Rafael Harpaz, Mark J Papania. Can a Minimum Rate of Investigation of Measleslike Illnesses Serve as a Standard for Evaluating Measles Surveillance[J]. Journal of Infectious Diseases,2004, 189(Suppl 1): S204-209. [ 6] 广东省卫生厅.广东省麻疹疫情调查处理技术规范(试行)[Z]. 2013-07-25. [ 7]黄 晖,阮 峰,关天姬,等.珠海市2005-2011年麻疹减毒活疫苗强化免疫前后麻疹流行病学特征[J].实用预防医学，2013,20（2）:183-184. [ 8]叶中文,黄晖,晁斌,等.两起麻疹聚集性疫情的流行病学关联调查分析[J].实用预防医学，2014,21（9）:1077-1079. [ 9] Benn Sartorius,C Cohen,T Chirwa,et al. Identifying high-risk areas for sporadic measles outbreaks: lessons from South Africa[J]. Bull World Health Organ.2013, 91(3): 174–183. [10 ] 龙腾锐,郭劲松,霍国友,等.水质综合评价的Hopfield网络模型[J]. 重庆建筑大学学报.2002,24(2):57-59. [11 ] 陈仕鸿,刘晓庆.基于离散型Hopfield神经网络的台风灾情评估模型[J].自然灾害学报，2011,20（5）:47-52. [12 ]World Health Organization. Measles Vaccines: WHO Position Paper [J]. WER,2009, 84 (35):349-360. [13 ]World Health Organization. Monitoring progress towards measles elimination[J].WER,2010，85(49): 490-495.
2016, Vol. 23 
