SEIHRS_gv model——predicting the influenza-like illness epidemic trend based on short term data
-
摘要:
目的 利用天津市流行性感冒(简称流感)样病例(influenza-like illness, ILI)监测数据，开发ILI疫情流行趋势预测模型；量化评估疫情防控措施对ILI产生的医疗负担影响。 方法 选取2023年11月6日―2023年11月15日天津市ILI数据进行SEIHRS_gv模型拟合，以2023年11月15日―2024年3月31日数据进行模型验证。选择均方根误差(root mean square error, RMSE)、平均绝对误差(mean absolute error, MAE)及决定系数(r-square, R2)评价模型预测能力。 结果 SEIHRS_gv模型可预测ILI疫情流行趋势、峰值及拐点，使用10 d数据进行预测，当R2达到0.85，RMSE为949.5，提高疫情防控措施强度可减少就诊人群数量。 结论 SEIHRS_gv模型在本轮ILI疫情预测中仅需几天数据，就可获得高精度预测结果，可作为评估医院就诊压力、指导疫情控制措施的高效预测模型。 Abstract:Objective To develop a prediction model for the epidemic trend of influenza-like illness using monitoring data from Tianjin City and quantitatively evaluate the impact of epidemic prevention and control measures on the medical burden caused by influenza-like illness. Methods The data from November 6, 2023 to November 15, 2023 were used for fitting the SEIHRS_gv model, and the data from November 15, 2023 to March 31, 2024 were using for validating. Root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination r-square (R2) were used to evaluate the predictive ability of the model. Results The SEIHRS_gv model could predict the trend, peak, and crucial point of influenza-like illness epidemics. Using 10 days of data for prediction, with an R2 of 0.85 and an RMSE of 949.5. Increasing the intensity of epidemic prevention and control measures could reduce the number of patients seeking medical treatment. Conclusions The SEIHRS_gv model required a few days of data for prediction in this round of influenza-like illness epidemic prediction and had high accuracy in prediction results, which could serve as an efficient predictive model to evaluate the pressure of hospital visits and guide the implementation intensity of epidemic control measures. -
Key words:
- Influenza-like illness /
- SEIR model /
- Epidemic prevention and control
-
图 1 每日新增确诊病例数影响因素的相关性分析表
Figure 1. Correlation analysis of factors affecting the daily number of newly confirmed cases
图 3 每日主动就诊人数预测曲线和实际曲线对比
Figure 3. Comparison between the predicted curve and the actual curve of the daily visits
图 4 每日主动就诊人数预测曲线和实际曲线对比(7、8、9 d)
Figure 4. Comparison between the predicted curve and the actual curve of the daily visits(7, 8, 9 d)
图 5 不同政策收缩指数取值的每日就诊人数曲线
Figure 5. Curves of daily visits with different values of the policy contraction index
图 6 不同α取值下的每日就诊人数预测曲线
Figure 6. Prediction curves of daily visits in different alpha values
表 1 SEIHRS_gv模型参数取值表
Table 1. Parameter values of the SEIHRS_gv model
参数 Parameter 定义 Definition 取值 Value 来源 Source β 传播系数 Infectious rate 0.045 MCMC α 单位时间内，潜伏期人群以一定的速率α转化为感染人群
Within a unit of time, latent individuals transform into infected individuals at a rate of α1、1/10、1/20、1/50 实际疫情推算
Actual Epidemicc 感染人群每天接触人数
Number of daily contacts of infected individuals10 文献
Literature[9-10]pH 感染人群以一定比例pH需要去医院就诊
Infected individuals need to seek medical attention at a rate of pH0.001 MCMC pHR 主动就诊人群以一定比例pHR恢复健康
Proactively seeking medical treatment leads to a recovery rate of pHR for the population0.999 实际疫情推算
Actual epidemicγH 假设感染人群从感染到需要去医院就诊的时间为t，γH为t的倒数
The reciprocal of t, which t is the time from infection to the need for medical treatment for infected individuals1/2 文献
Literature[9, 11-12]γI 单位时间内从患者群体中移除的比率为γI，设置为发病到康复的时间的倒数
γI is the removal rate from the patient population per unit time, it was set up as the reciprocal of the time from onset to recovery1/5 文献
Literature[13-14]γHR 假设感染人群从去医院就诊到康复的时间为t，γHR为t的倒数
Assuming that t is the time that an infected person needs to go to the hospital for treatment, γHR is the reciprocal of t1/5 文献
Literature[15-16]ρ 已康复人群由于体内抗体水平下降或免疫逃逸失去部分免疫力并再次变得易感的比率
The rate that people with immunity from ILI became susceptible again due to a decrease in antibody levels or immune evasion1/90 实际疫情推算
Actual epidemicgv 政策收缩指数[8] Government stringency index 0~1 文献 Literature[8] 注：MCMC，马尔科夫链蒙特卡洛; ILI, 流行性感冒样病例。
Note: MCMC, Markov chain Monte Carlo; ILI, influenza-like illness.
下载: 导出CSV
表 2 不同政策收缩指数取值对应措施
Table 2. Corresponding measures for different policy contraction index values
收缩指数
Stringency index疫情防控措施
Anti-pandemic measures0 无政策 No policy 0.25 提倡个人防护、增大宣传活动、倡导减缓流动，有症状人自愿检测
Promote personal protection, increase publicity activities, advocate slowing down mobility, and encourage symptomatic individuals to voluntarily undergo testing0.50 限制公众集会、公共场所关闭、取消公共活动，有症状的人进行检测，在特定公开场合佩戴口罩
Restricting public gatherings, closing public places, canceling public activities, testing symptomatic individuals, and wearing masks in specific public places0.75 学校关闭、工作场所关闭，区域筛查，在公共场所佩戴口罩
School and workplace closure, regional screening and wearing masks in public places1.00 居家要求，全员筛查，严格口罩佩戴
Home requirements, full staff screening and strict mask wearing
下载: 导出CSV
-
[1] 中华人民共和国卫生部办公厅. 卫生部办公厅关于印发流感样病例暴发疫情处置指南(2012年版)的通知[EB/OL]. (2012-11-05)[2024-04-30]. https://www.gov.cn/gzdt/2012-11/13/content_2264683.htm .[2] 曹滋莲, 郑雅旭, 杨娟, 等. 2017—2020年上海市流感相关流感样病例门急诊疾病负担估计[J]. 国际病毒学杂志, 2023, 30(3): 214-219. DOI: 10.3760/cma.j.issn.1673-4092.2023.03.008.Cao ZL, Zheng YX, Yang J, et al. Disease burden of influenza-associated influenza-like illness in outpatients and emergency departments in Shanghai, 2017-2020[J]. Int J Virol, 2023, 30(3): 214-219. DOI: 10.3760/cma.j.issn.1673-4092.2023.03.008. [3] 侯世娇, 闫红静, 甄珊珊, 等. 2010—2021年上海市闵行区流感样病例及病原学特征[J]. 中华微生物学和免疫学杂志, 2022, 42(12): 965-972. DOI: 10.3760/cma.j.cn112309-20220711-00231.Hou SJ, Yan HJ, Zhen SS, et al. Influenza-like illnesses and their etiological characteristics in Minhang District of Shanghai during 2010 to 2021[J]. Chin J Microbiol Immunol, 2022, 42(12): 965-972. DOI: 10.3760/cma.j.cn112309-20220711-00231. [4] 马春娜, 吴双胜, 张莉, 等. 北京市2015—2020年流感流行季流感样病例和流感病原学分析[J]. 中华实验和临床病毒学杂志, 2021, 35(1): 44-48. DOI: 10.3760/cma.j.cn112866-20200803-00220.Ma CN, Wu SS, Zhang L, et al. Analysis of the influenza surveillance in Beijing during 2015-2020 influenza seasons[J]. Chin J Exp Clin Virol, 2021, 35(1): 44-48. DOI: 10.3760/cma.j.cn112866-20200803-00220. [5] Feng LZ, Feng S, Chen T, et al. Burden of influenza-associated outpatient influenza-like illness consultations in China, 2006-2015: a population-based study[J]. Influenza Other Respir Viruses, 2020, 14(2): 162-172. DOI: 10.1111/irv.12711. [6] World Health Organization. COVID-19 eliminated a decade of progress in global level of life expectancy[EB/OL]. (2024-05-24)[2024-06-11]. https://www.who.int/news/item/24-05-2024-covid-19-eliminated-a-decade-of-progress-in-global-level-of-life-expectancy .[7] 天津市统计局. 天津统计年鉴[EB/OL] (2023-11-23)[2024-04-30]. https://stats.tj.gov.cn/tjsj_52032/tjnj/ .[8] 金鑫, 马钰淇, 张寒松, 等. 应对突发公共卫生事件的对策分析——以天津市为例[J]. 中国公共卫生管理, 2024, 40(1): 42-45. DOI: 10.19568/j.cnki.23-1318.2024.01.0010.Jin X, Ma YQ, Zhang HS, et al. Analysis of countermeasures in response to public health emergencies: taking Tianjin as an example[J]. Chin J PHM, 2024, 40(1): 42-45. DOI: 10.19568/j.cnki.23-1318.2024.01.0010. [9] Cai J, Deng XW, Yang J, et al. Modeling transmission of SARS-CoV-2 Omicron in China[J]. Nat Med, 2022, 28(7): 1468-1475. DOI: 10.1038/s41591-022-01855-7. [10] Mistry D, Litvinova M, Pastore Y Piontti A, et al. Inferring high-resolution human mixing patterns for disease modeling[J]. Nat Commun, 2021, 12(1): 323. DOI: 10.1038/s41467-020-20544-y. [11] 李雪辉, 吕楠, 陈杭薇, 等. 229例成人甲型流感患者临床特征分析[J]. 国际呼吸杂志, 2011, 31(14): 1041-1043. DOI: 10.3760/cma.j.issn.1673-436X.2011.014.001.Li XH, Lyu N, Chen HW, et al. Clinical features of 229 cases of adults influenza[J]. Int J Respir, 2011, 31(14): 1041-1043. DOI: 10.3760/cma.j.issn.1673-436X.2011.014.001. [12] 彭伟, 董靖, 王烁. 成人乙型流感病毒肺炎的临床特点及影像学特征[J]. 中国医师杂志, 2023, 25(2): 170-172. DOI: 10.3760/cma.j.cn431274-20230131-00071.Peng W, Dong J, Wang S. Clinical and imaging features of adult influenza B viral pneumonia[J]. J Chin Physician, 2023, 25(2): 170-172. DOI: 10.3760/cma.j.cn431274-20230131-00071. [13] 张巧利, 钟新光, 刘志权, 等. 甲型H1N1流行性感冒56例流行病学和临床学分析[J]. 中华传染病杂志, 2009, 27(10): 586-590. DOI: 10.3760/cma.j.issn.1000-6680.2009.10.003.Zhang QL, Zhong XG, Liu ZQ, et al. Analysis of epidemiological and clinical characteristics of 56 confirmed cases of influenza a(H1N1)[J]. Chin J Infect Dis, 2009, 27(10): 586-590. DOI: 10.3760/cma.j.issn.1000-6680.2009.10.003. [14] 刘映霞, 杨大国, 谢靖婧, 等. 深圳市甲型H1N1流行性感冒确诊病例流行病学及临床特征[J]. 中华传染病杂志, 2009, 27(10): 582-585. DOI: 10.3760/cma.j.issn.1000-6680.2009.10.002.Liu YX, Yang DG, Xie JJ, et al. Epidemiology and clinical manifestations of confirmed cases of influenza a(H1N1)in Shenzhen[J]. Chin J Infect Dis, 2009, 27(10): 582-585. DOI: 10.3760/cma.j.issn.1000-6680.2009.10.002. [15] 刘博, 李素英. 2023年北京某高校流感流行季甲型流感感染及治疗状况分析[J]. 国际病毒学杂志, 2023, 30(5): 419-422. DOI: 10.3760/cma.j.issn.1673-4092.2023.05.016.Liu B, Li SY. Infection status and treatment of influenza A in one university in Beijing during the influenza season of 2023[J]. Int J Virol, 2023, 30(5): 419-422. DOI: 10.3760/cma.j.issn.1673-4092.2023.05.016. [16] 李亚品, 钱全, 方立群, 等. 中国大陆2009年早期确诊的420例甲型H1N1流感病例流行病学特征分析[J]. 中华流行病学杂志, 2009, 30(11): 1102-1105. DOI: 10.3760/cma.j.issn.0254-6450.2009.11.002.Li YP, Qian Q, Fang LQ, et al. Epidemiological characteristics of 420 influenza A (H1N1) cases confirmed in the early stage of the epidemic in mainland China[J]. Chin J Epidemiol, 2009, 30(11): 1102-1105. DOI: 10.3760/cma.j.issn.0254-6450.2009.11.002. [17] World Health Organization. Influenza (seasonal)[EB/OL]. (2023-10-03)[2024-04-30]. https://www.who.int/news-room/fact-sheets/detail/influenza-(seasonal) .[18] Nair H, Brooks WA, Katz M, et al. Global burden of respiratory infections due to seasonal influenza in young children: a systematic review and meta-analysis[J]. Lancet, 2011, 378(9807): 1917-1930. DOI: 10.1016/S0140-6736(11)61051-9. [19] Thompson WW, Weintraub E, Dhankhar P, et al. Estimates of US influenza-associated deaths made using four different methods[J]. Influenza Other Respir Viruses, 2009, 3(1): 37-49. DOI: 10.1111/j.1750-2659.2009.00073.x. [20] Shin HY. A multi-stage SEIR(D) model of the COVID-19 epidemic in Korea[J]. Ann Med, 2021, 53(1): 1159-1169. DOI: 10.1080/07853890.2021.1949490. [21] Girardi P, Gaetan C. An SEIR model with time-varying coefficients for analyzing the SARS-CoV-2 epidemic[J]. Risk Anal, 2023, 43(1): 144-155. DOI: 10.1111/risa.13858. [22] Mwalili S, Kimathi M, Ojiambo V, et al. SEIR model for COVID-19 dynamics incorporating the environment and social distancing[J]. BMC Res Notes, 2020, 13(1): 352. DOI: 10.1186/s13104-020-05192-1. [23] Hou YZ, Bidkhori H. Multi-feature SEIR model for epidemic analysis and vaccine prioritization[J]. PLoS One, 2024, 19(3): e0298932. DOI: 10.1371/journal.pone.0298932. [24] Khairulbahri M. The SEIR model incorporating asymptomatic cases, behavioral measures, and lockdowns: lesson learned from the COVID-19 flow in Sweden[J]. Biomed Signal Process Control, 2023, 81: 104416. DOI: 10.1016/j.bspc.2022.104416. [25] Guan DB, Wang DP, Hallegatte S, et al. Global supply-chain effects of COVID-19 control measures[J]. Nat Hum Behav, 2020, 4(6): 577-587. DOI: 10.1038/s41562-020-0896-8. [26] Heredia Cacha I, Sáinz-Pardo Díaz J, Castrillo M, et al. Forecasting COVID-19 spreading through an ensemble of classical and machine learning models: Spain's case study[J]. Sci Rep, 2023, 13(1): 6750. DOI: 10.1038/s41598-023-33795-8. [27] Durai CAD, Begum A, Jebaseeli J, et al. COVID-19 pandemic, predictions and control in Saudi Arabia using SIR-F and age-structured SEIR model[J]. J Supercomput, 2022, 78(5): 7341-7353. DOI: 10.1007/s11227-021-04149-w. [28] Abdul Salam M, Taha S, Ramadan M. COVID-19 detection using federated machine learning[J]. PLoS One, 2021, 16(6): e0252573. DOI: 10.1371/journal.pone.0252573. [29] Gothai E, Thamilselvan R, Rajalaxmi RR, et al. Prediction of COVID-19 growth and trend using machine learning approach[J]. Mater Today Proc, 2023, 81: 597-601. DOI: 10.1016/j.matpr.2021.04.051. [30] Ma YF, Xu SJ, An Q, et al. Coronavirus disease 2019 epidemic prediction in Shanghai under the "dynamic zero-COVID policy" using time-dependent SEAIQR model[J]. J Biosaf Biosecur, 2022, 4(2): 105-113. DOI: 10.1016/j.jobb.2022.06.002. [31] 王小莉, 曹志冬, 曾大军, 等. 应用SEIR模型预测2009年甲型H1N1流感流行趋势[J]. 国际病毒学杂志, 2011, 18(6): 161-165. DOI: 10.3760/cma.j.issn.1673-4092.2011.06.001.Wang XL, Cao ZD, Zeng DJ, et al. Prediction of the epidemic trends of pandemic H1N1 2009, using SEIR model[J]. Int J Virol, 2011, 18(6): 161-165. DOI: 10.3760/cma.j.issn.1673-4092.2011.06.001. [32] 刘晓曼, 郭丽茹, 孔梅, 等. 天津市520例儿童流感样病例病原谱以及流行病学特征分析[J]. 中华微生物学和免疫学杂志, 2022, 42(12): 973-980. DOI: 10.3760/cma.j.cn112309-20220922-00308.Liu XM, Guo LR, Kong M, et al. Analysis of the pathogenic spectrum and epidemiological characteristics of 520 children with influenza-like illnesses in Tianjin[J]. Chin J Microbiol Immunol, 2022, 42(12): 973-980. DOI: 10.3760/cma.j.cn112309-20220922-00308. -
点击查看大图
图(6) / 表(2)
计量
- 文章访问数: 28
- HTML全文浏览量: 13
- PDF下载量: 14
- 被引次数: 0