DLNM和LSTM神经网络对临沂市手足口病发病的预测效果比较

doi:10.6040/j.issn.1671-7554.0.2021.0707

摘要/Abstract

摘要： 目的运用分布滞后非线性模型(DLNM)和长短期记忆(LSTM)神经网络对山东省临沂市手足口病(HFMD)发病趋势进行分析和预测,为该病的有效防控提供参考依据。方法对临沂市2011年1月1日至2015年12月31日HFMD日发病数据分别进行DLNM和LSTM神经网络建模拟合,以2016年1月1日至2017年12月31日发病数据检验并比较两模型的预测效果。结果 2011年1月1日至2017年12月31日临沂市共报告HFMD 25 999例。DLNM和LSTM神经网络外推预测2016年1月1日至2017年12月31日发病数的均方根误差(RMSE)分别为11.93和5.74,平均绝对误差(MAE)分别为7.93和3.60,提示LSTM神经网络的预测精度优于DLNM,预测结果与实际情况基本一致。结论 LSTM神经网络对临沂市HFMD发病趋势的拟合程度和预测效果较好,可为该病的预测预警提供指导。

关键词: 手足口病, 预测, 分布滞后非线性模型, 长短期记忆神经网络

Abstract: Objective To analyze and predict the incidence trend of hand, foot and mouth disease(HFMD)in Linyi City, Shandang Province by using the distributed lag non-linear model(DLNM)and long-short term memory(LSTM)neural network, and to provide reference for effective prevention and control of the disease. Methods The daily incidence data from Jan. 1, 2011 to Dec. 31, 2015 were collected to establish the DLNM and LSTM neural network, respectively. The daily incidence data from Jan. 1, 2016 to Dec. 31, 2017 were used to test and compare the prediction effects of the two models. Results A total of 25,999 HFMD cases were reported during Jan. 1, 2011 to Dec. 31, 2017. The root mean square error(RMSE)of DLNM and LSTM neural network extrapolation prediction from Jan. 1, 2016 to Dec. 31, 2017 were 11.93 and 5.74, respectively, and the mean absolute deviation(MAE)were 7.93 and 3.60, respectively, indicating the prediction accuracy of LSTM was better than that of DLNM, and the prediction results were basically consistent with the actual situation. Conclusion LSTM neural network has a good fitting and prediction effect on the incidence trend of HFMD in Linyi City, which can provide guidance for the prediction and warning of the disease.

Key words: Hand, foot and mouth disease, Prediction, Distributed lag non-linear model, Long-short term memory neural network

中图分类号:

R181.3

冯一平,孙大鹏,王显军,纪伊曼,刘云霞. DLNM和LSTM神经网络对临沂市手足口病发病的预测效果比较[J]. 山东大学学报 (医学版), 2022, 60(2): 96-101.

FENG Yiping, SUN Dapeng, WANG Xianjun, JI Yiman, LIU Yunxia. Comparison of prediction effects of DLNM and LSTM neural network on the incidence of hand, foot and mouth disease in Linyi City[J]. Journal of Shandong University (Health Sciences), 2022, 60(2): 96-101.

参考文献

[1] Guerra AM, Orille E, Waseem M. Hand foot and mouth disease [DB/OL].(2021-09-20)[2021-04-27]. https://www.ncbi.nlm.nih.gov/books/NBK431082/.
[2] 崔金朝, 聂陶然, 任敏睿, 等. 2008-2018年中国5岁及以下儿童手足口病死亡病例流行病学特征[J]. 中华流行病学杂志, 2020, 41(7): 1041-1046. CUI Jinchao, NIE Taoran, REN Minrui, et al. Epidemiological charact-eristics of fatal cases of hand, foot, and mouth disease in children under 5 years old in China, 2008-2018 [J]. Chinese Journal of Epidemiology, 2020, 41(7): 1041-1046.
[3] 《手足口病诊疗指南(2018版)》编写专家委员会. 手足口病诊疗指南(2018年版)[J].中华传染病杂志,2018, 36(5): 257-263. Expert Committee for Guidelines for the diagnosis and treatment of hand, foot, and mouth disease. Guidelines for the diagnosis and treatment of hand, foot, and mouth disease(2018 Edition)[J]. Chinese Journal of Infectious Diseases, 2018, 36(5): 257-263.
[4] He X, Zhang M, Zhao C, et al. From monovalent to multivalent vaccines, the exploration for potential preventive strategies against hand, foot, and mouth disease(HFMD)[J]. Virol Sin, 2021, 36(2): 167-175.
[5] Fang CY, Liu CC. Recent development of Enterovirus a vaccine candidates for the prevention of hand, foot, and mouth disease [J]. Expert Rev Vaccines, 2018, 17(9): 819-831.
[6] Liu Z, Meng Y, Xiang H, et al. Association of short-term exposure to meteorological factors and risk of hand, foot, and mouth disease: a systematic review and meta-analysis [J]. Int J Environ Res Public Health, 2020, 17(21): 8017.
[7] Wu X, Hu S, Kwaku AB, et al. Spatio-temporal clustering analysis and its determinants of hand, foot and mouth disease in Hunan, China, 2009-2015 [J]. BMC Infect Dis, 2017, 17(1): 645.
[8] Gasparrini A. Distributed lag linear and non-linear models in R: the package dlnm [J]. J Stat Softw, 2011, 43(8): 1-20.
[9] Gasparrini A, Armstrong B, Kenward MG. Multivariate meta-analysis for non-linear and other multi-parameter associations [J]. Stat Med, 2012, 31(29): 3821-3839.
[10] Cheng Q, Bai L, Zhang Y, et al. Ambient temperature, humidity and hand, foot, and mouth disease: a systematic review and meta-analysis [J]. Sci Total Environ, 2018, 625: 828-836. doi: 10.1016/j.scitotenv.2018.01.006.
[11] Chen S, Liu X, Wu Y, et al. The application of meteorological data and search index data in improving the prediction of HFMD: a study of two cities in Guangdong Province, China [J]. Sci Total Environ, 2019, 652: 1013-1021. doi: 10.1016/j.scitotenv.2018.10.304.
[12] Gu S, Huang R, Yang J, et al. Exposure-lag-response association between sunlight and schizophrenia in Ningbo, China [J]. Environ Pollut, 2019, 247: 285-292. doi: 10.1016/j.envpol.2018.12.023.
[13] Gu J, Liang L, Song H, et al. A method for hand-foot-mouth disease prediction using GeoDetector and LSTM model in Guangxi, China [J]. Sci Rep, 2019, 9(1): 17928. doi:10.1038/s41598-019-54495-2.
[14] Xu X, Yoneda M. Multitask air-quality prediction based on LSTM-autoencoder model [J]. IEEE Trans Cybern, 2021, 51(5): 2577-2586.
[15] Rahman M, Islam D, Mukti RJ, et al. A deep learning approach based on convolutional LSTM for detecting diabetes [J]. Comput Biol Chem, 2020, 88: 107329. doi: 10.1016/j.compbiolchem.2020.107329.
[16] Liao J, Qin Z, Zuo Z, et al. Spatial-temporal mapping of hand foot and mouth disease and the long-term effects associated with climate and socio-economic variables in Sichuan Province, China from 2009 to 2013 [J]. Sci Total Environ, 2016, 563/564: 152-159. doi: 10.1016/j.scitotenv.2016.03.159.
[17] Yang F, Ma Y, Liu FF, et al. Short-term effects of rainfall on childhood hand, foot and mouth disease and related spatial heterogeneity: evidence from 143 cities in mainland China [J]. BMC Public Heal, 2020, 20(1): 1528.
[18] 祁海萍. 2009-2013年甘肃省三市手足口病报告病例流行特征与趋势分析[D]. 兰州: 兰州大学, 2015.
[19] 刘涛, 王显军, 姜宝法, 等. SARIMA模型预测山东省手足口病发病趋势[J]. 中国卫生统计, 2013, 30(5): 697-700. LIU Tao, WANG Xianjun, JIANG Baofa, et al. Prediction of hand-foot-mouth disease incidence using SARIMA model in Shandong Province [J]. Chinese Journal of Health Statistics, 2013, 30(5): 697-700.
[20] 党德建, 朱杰辉, 张超, 等. 炎症因子在手足口病重症和重症天数判断中的应用[J]. 现代预防医学, 2017, 44(20): 3817-3821. DANG Dejian, ZHU Jiehui, ZHANG Chao, et al. Application of the inflammatory factor in determination of severe hand-foot-and-mouth disease cases and duration [J]. Modern Preventive Medicine, 2017, 44(20): 3817-3821.
[21] 马晓梅, 徐学琴, 闫国立, 等. BP神经网络和决策树分析在重症手足口病临床早期预警指标中的应用[J]. 中国卫生统计, 2019, 36(3): 381-383.
[22] 段清浩,李廷荣,杨连建.重庆市沙坪坝区手足口病发病数预测模型的构建与评价[J]. 西南国防医药, 2018, 28(10): 979-981. DUAN Qinghao, LI Tingrong, YANG Lianjian. Construction and evaluation of prediction model of hand foot and mouth disease incidence in Shapingba District of Chongqing [J]. Medical Journal of National Defending Forces in Southwest China, 2018, 28(10): 979-981.
[23] Xiang J, Hansen A, Liu Q, et al. Association between dengue fever incidence and meteorological factors in Guangzhou, China, 2005-2014 [J]. Environ Res, 2017, 153: 17-26. doi: 10.1016/j.envres.2016.11.009.
[24] Gao Y, Niu Y, Sun W, et al. Climate factors driven typhus group rickettsiosis incidence dynamics in Xishuangbanna Dai autonomous prefecture of Yunnan Province in China, 2005-2017 [J]. Environ Health, 2020, 19(1): 3.
[25] Pathan RK, Biswas M, Khandaker MU. Time series prediction of COVID-19 by mutation rate analysis using recurrent neural network-based LSTM model [J]. Chaos Solitons Fractals, 2020, 138: 110018. doi: 10.1016/j.chaos.2020.110018.
[26] Guo Y, Feng Y, Qu F, et al. Prediction of hepatitis E using machine learning models [J]. PLoS One, 2020, 15(9): e0237750. doi: 10.1371/journal.pone.0237750.
[27] 祝寒松, 陈思, 王明斋, 等. 厦门市2013-2017年手足口病发病与气象因素影响分析[J]. 中华流行病学杂志, 2019, 40(5): 531-536. ZHU Hansong, CHEN Si, WANG Mingzhai, et al. Analysis on association between incidence of hand foot and mouth disease and meteorological factors in Xiamen, 2013-2017 [J]. Chinese Journal of Epidemiology, 2019, 40(5): 531-536. doi: 10.3760/cma.j.issn.0254-6450.2019.05.008.
[28] Lu J, Bu P, Xia X, et al. A new deep learning algorithm for detecting the lag effect of fine particles on hospital emergency visits for respiratory diseases [J]. IEEE Access, 2020, 8: 145593-600. doi:https://doi.org/10.1109/ACCESS.2020.3013543.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed