Research on predictive analysis methods for symptom monitoring based on time series and spatio-temporal aggregation detection
DUAN Wei and ZHOU Xiaofang contributed equally to this article
More Information
-
摘要:
目的 探索基于时间序列和时空聚集性探测的症状监测预测方法，为有效分析和利用症状监测数据提供参考依据。 方法 采用差分自回归移动平均(autoregressive integrated moving average, ARIMA)模型和Holt-Winters模型进行时间序列分析，通过回顾性时空聚集性探测进行聚集区域和时间综合探测分析。 结果 以在X市监测为例，2023年1月1日―2023年4月30日共监测到34 207人次出现与传染病相关症状并前往医疗机构就诊。4月1日―4月19日的模型预测值与实际监测值比较发现，Holt-Winters模型对数据的预测情况优于ARIMA模型，误差更小，几乎所有实际值均在预测值95% CI内。时空扫描分析结果显示，某市就诊人群中具有所监测症状者居住社区涵盖该市9个街道，发热、咳嗽、腹痛和头痛的1类聚集地为D街道、E街道、F街道、G街道和A街道；咽痛和恶心的1类聚集地为D街道、A街道和G街道；腹泻和呕吐的1类集聚地为G街道和D街道。所监测症状的发生时间主要集中在2022年12月―2023年4月。 结论 Holt-Winters模型对症状数据具有较好的时间趋势预测效果，通过对症状监测数据进行时空扫描分析可及时发现传染病的聚集情况，为防控工作提供重要的空间、时间和时空联合指示。症状监测数据可被用于监测传染病的潜在流行情况，为实现早期预警提供参考。 Abstract:Objective To explore appropriate prediction methods for symptom monitoring based on time series and spatio-temporal aggregation detection, and to provide a foundational reference for the effective analysis and utilization of symptom monitoring data. Methods In this study, ARIMA and Holt-Winters models were used for time-series analysis, and aggregation regions and time-integrated detection analyses were conducted through a retrospective temporal-spatial clustering detection analysis. Results Taking the surveillance in X City as an example, from January 1 to April 30, 2023, 34 207 individuals were monitored for symptoms that may be associated with communicable diseases and visited a healthcare facility. Comparing the model predicted value between April 1 to April 19 with the actual monitoring value found that the Holt-Winters model predicted the data better than the ARIMA model with smaller errors, and almost all the actual values were within the 95% confidence interval of the predicted value. The spatio-temporal scan analysis showed that the residential community of patients with monitored symptoms in the certain city covered 9 clusters of the city. Category 1 included D street, E street, A street, F street, G street and A street; category 1 cluster of sore throat and nausea was D street, A street and G street; category 1 cluster of diarrhea and vomiting was G street and D street. The onset of the monitored symptoms was mainly concentrated between December 2022 and April 2023. Conclusions The Holt-Winters model has a good time trend prediction effect on symptom data, and the aggregation of infectious diseases can be detected in time by analyzing symptom surveillance data through spatio-temporal scanning, which provides important joint spatial, temporal and spatio-temporal indications for prevention and control efforts. Symptom surveillance data can be used to monitor the potential prevalence of infectious diseases and provide reference for realizing early warning. -
图 1 时间序列、ACF、PACF图
ACF：自相关函数；PACF：偏自相关函数。
Figure 1. Time series, ACF, and PACF graphs
ACF: autocorrelation function; PACF: partial autocorrelation function.
图 2 ARIMA和Holt-Winters建模比较
A：ARIMA模型；B：Holt-Winters模型。
Figure 2. Comparison of ARIMA and Holt-Winters modeling
A: autoregressive integrated moving average model; B: Holt-Winters model.
图 3 2022年8月―2023年9月某市所监测症状时间分布图
2023年5月未收集到数据。
Figure 3. Temporal distribution of monitored symptoms in a city from August 2022 to September 2023
No date were collected in May 2023.
表 1 某市9个街道所监测症状发生率
Table 1. Incidence of symptoms monitored in 9 streets of a city
地区Area 监测症状发生率Incidence of monitored symptoms/% 发热
Fever咳嗽
Cough咽痛
Sore
throat腹泻
Diarrhoea腹痛
Abdominal pain恶心
Nausea呕吐
Vomiting头痛
HeadacheA街道A street 1.04 1.04 0.10 0.30 1.22 0.11 0.16 0.36 B街道B street 0.36 0.25 0.04 0.09 0.36 0.03 0.05 0.10 C街道C street 1.23 1.17 0.08 0.43 2.12 0.13 0.23 0.49 D街道D street 3.68 2.19 0.12 0.81 3.43 0.30 0.65 0.92 E街道E street 1.95 1.26 0.06 0.53 2.60 0.06 0.19 0.74 F街道F street 0.51 0.32 0.02 0.08 0.59 0.02 0.07 0.09 G街道G street 1.65 1.04 0.08 0.49 1.86 0.11 0.36 0.39 H街道H street 1.23 0.56 0.03 0.24 0.97 0.01 0.18 0.22 I街道I street 0.69 0.35 0.03 0.22 0.74 0.05 0.13 0.19 
下载: 导出CSV
表 2 2022年8月―2023年9月某市所监测症状的时空扫描分析
Table 2. Spatio-temporal scanning analysis with monitored symptoms in a city from August 2022 to September 2023
症状
Symptom聚集区类型
Cluster type聚集区域
Cluster area聚集时间/(年-月)
Cluster time
(year-month)实际发病数
Number of cases理论发病数
Expected casesLLR值
valueRR值
valueP值
value发热Fever 1 D街道、E街道、F街道、G街道、A街道
D street, E street, F street, G street, A street2022-11―2023-04 1 732 915.95 383.57 2.42 < 0.05 2 H街道H street 2022-11―2022-12 170 59.48 69.34 2.93 < 0.05 3 C街道C street 2023-03 85 21.47 53.86 4.01 < 0.05 咳嗽Cough 1 D街道、E街道、F街道、G街道、A街道
D street, E street, F street, G street, A street2022-12―2023-01 1 021 237.60 806.82 5.64 < 0.05 2 C街道C street 2022-12―2023-04 229 79.21 96.63 3.02 < 0.05 3 H街道H street 2022-12―2023-01 95 45.79 20.48 2.10 < 0.05 咽痛Sore throat 1 D街道、A街道、G街道D street, A street, G street 2022-11―2023-01 86 27.41 46.48 3.99 < 0.05 腹泻Diarrhoea 1 G街道、D街道G street, D street 2023-03―2023-09 170 57.01 78.10 3.28 < 0.05 2 C街道C street 2023-01―2023-07 95 40.59 27.85 2.45 < 0.05 3 I街道、A街道I street, A street 2023-03―2023-04 154 90.82 19.84 1.79 < 0.05 腹痛Abdominal pain 1 E街道、F街道、G街道、D街道、A街道
E street, F street, G street, D street, A street2023-03―2023-04 723 347.31 169.21 2.25 < 0.05 2 C街道C street 2023-03―2023-07 309 119.23 108.04 2.69 < 0.05 3 H街道H street 2023-03―2023-04 110 66.93 11.76 1.66 < 0.05 恶心Nausea 1 D街道、A街道、G街道D street, A street, G street 2022-12―2023-04 105 55.82 21.15 2.22 < 0.05 2 C街道C street 2022-11―2023-03 22 8.44 7.77 2.71 < 0.05 呕吐Vomiting 1 G街道、D街道G street, D street 2023-03―2023-07 95 24.08 62.99 4.37 < 0.05 2 C街道、I街道C street, I street 2023-03―2023-04 58 24.53 17.22 2.48 < 0.05 头痛Headache 1 E街道、F街道、G街道、D街道、A街道
E street, F street, G street, D street, A street2022-10―2023-04 464 316.09 41.07 1.71 < 0.05 2 C街道C street 2023-06―2023-09 56 24.90 14.65 2.30 < 0.05
下载: 导出CSV
-
[1] Rader B, Scarpino SV, Nande A, et al. Crowding and the shape of COVID-19 epidemics[J]. Nat Med, 2020, 26(12): 1829-1834. DOI: 10.1038/s41591-020-1104-0. [2] Chen H, Zeng D, Yan P. Infectious disease informatics syndrome surveillance for public health and bio-defense[M]. Beijing Science Press, 2011: 1-9. [3] Papadomanolakis-Pakis N, Maier A, van Dijk A, et al. Development and assessment of a hospital admissions-based syndromic surveillance system for COVID-19 in Ontario, Canada: ACES Pandemic Tracker[J]. BMC Public Health, 2021, 21(1): 1230. DOI: 10.1186/s12889-021-11303-9. [4] Güemes A, Ray S, Aboumerhi K, et al. A syndromic surveillance tool to detect anomalous clusters of COVID-19 symptoms in the United States[J]. Sci Rep, 2021, 11(1): 4660. DOI: 10.1038/s41598-021-84145-5. [5] Daisuke Y, Yuta T, Takayuki K, et al. Large-scale epidemiological monitoring of the COVID-19 epidemic in Tokyo[J]. Lancet Reg Health, 2020, 3: 100016. DOI: 10.1016/j.lanwpc.2020.100016. [6] Mahmud AS, Chowdhury S, Sojib KH, et al. Participatory syndromic surveillance as a tool for tracking COVID-19 in Bangladesh[J]. Epidemics, 2021, 35: 100462. DOI: 10.1016/j.epidem.2021.100462. [7] Fulcher IR, Boley EJ, Gopaluni A, et al. Syndromic surveillance using monthly aggregate health systems information data: methods with application to COVID-19 in Liberia[J]. Int J Epidemiol, 2021, 50(4): 1091-1102. DOI: 10.1093/ije/dyab094. [8] Gong MC, Liu L, Sun X, et al. Cloud-based system for effective surveillance and control of COVID-19: useful experiences from Hubei, China[J]. J Med Int Res, 2020, 22(4): e18948. DOI: 10.2196/18948. [9] Wang XY, Zhang Y, Gong WS. Analysis of the monitoring data of fever clinic symptoms during the pandemic of new coronavirus pneumonia in Xiangyang City[J]. J Public Health Prev Med, 2020, 31(4): 28-30. DOI: 10.1111/inr.12636. [10] Project TS. Assessment of syndromic surveillance in Europe[J]. Lancet, 2011, 378(9806): 1833-1834. DOI: 10.1016/S0140-6736(11)60834-9. [11] Hu WH, Sun HM, Wei YY, et al. Global infectious disease early warning models: an updated review and lessons from the COVID-19 pandemic[J]. Infect Dis Model, 2024, 10(2): 410-422. DOI: 10.1016/j.idm.2024.12.001. [12] Botz J, Wang DQ, Lambert N, et al. Modeling approaches for early warning and monitoring of pandemic situations as well as decision support[J]. Front Public Health, 2022, 10: 994949. DOI: 10.3389/fpubh.2022.994949. [13] Swapnarekha H, Behera HS, Nayak J, et al. Multiplicative Holts Winter model for trend analysis and forecasting of COVID-19 spread in India[J]. SN Computer Science, 2021, 2(5): 416. DOI: 10.1007/s42979-021-00808-0. [14] 马涛, 谢国祥, 孙红敏, 等. 2009―2016年南京市重症手足口病流行特征和时空聚集性分析[J]. 中华疾病控制杂志, 2018, 22(11): 1138-1143. DOI: 10.16462/j.cnki.zhjbkz.2018.11.011.Ma T, Xie GX, Sun HM et al. Analysis of epidemiological characteristics and temporal-spatial clustering analysis of severe hand-foot-mouth disease in Nanjing from 2009 to 2016[J]. Chin J Dis Control Prev, 2018, 22(11): 1138-1143. DOI: 10.16462/j.cnki.zhjbkz.2018.11.011. [15] Bagalkot SS, Dinesha HA, Naik N. Novel grey wolf optimizer based parameters selection for GARCH and ARIMA models for stock price prediction[J]. PeerJ Comput Sci, 2024, 10: e1735. DOI: 10.7717/peerj-cs.1735. [16] Lou HR, Wang X, Gao Y, et al. Comparison of ARIMA model, DNN model and LSTM model in predicting disease burden of occupational pneumoconiosis in Tianjin, China[J]. BMC Public Health, 2022, 22(1): 2167. DOI: 10.1186/s12889-022-14642-3. [17] 郑艳妮, 赵玉锐, 梁莉萍, 等. 基于SARIMA、Holt-Winters与Prophet三种时间序列模型的2021年武威市丙型病毒性肝炎发病预测[J]. 疾病预防控制通报, 2023, 38(5): 49-53, 68. DOI: 10.13215/j.cnki.jbyfkztb.2303005.Zheng YN, Zhao YR, Liang LP, et al. Forecasting the incidence of viral hepatitis C in Wuwei City in 2021 based on three time series models: SARIMA, Holt-Winters and Prophet[J]. Bull Dis Control Prev, 2023, 38(5): 49-53, 68. DOI: 10.13215/j.cnki.jbyfkztb.2303005. [18] Xian XB, Wang L, Wu XH, et al. Comparison of SARIMA model, Holt-Winters model and ETS model in predicting the incidence of foodborne disease[J]. BMC Infect Dis, 2023, 23(1): 803. DOI: 10.1186/s12879-023-08799-4. [19] Wang N, Zhuang WY, Ran Z, et al. Prediction of acute onset of chronic cor pulmonale: comparative analysis of Holt-Winters exponential smoothing and ARIMA model[J]. BMC Med Res Methodol, 2024, 24(1): 204. DOI: 10.1186/s12874-024-02325-z. [20] Deng LL, Han YJ, Wang JL, et al. Epidemiological characteristics of notifiable respiratory infectious diseases in Mainland China from 2010 to 2018[J]. Int J Environ Res Public Health, 2023, 20(5): 3946. DOI: 10.3390/ijerph20053946. [21] Si XH, Wang LP, Mengersen K, et al. Epidemiological features of seasonal influenza transmission among 11 climate zones in Chinese Mainland[J]. Infect Dis Poverty, 2024, 13(1): 4. DOI: 10.1186/s40249-024-01173-9. [22] 梅琳, 张弢, 胡弘, 等. 基于医疗机构的突发呼吸道传染病症状监测预警机制研究[J]. 中国医院管理, 2022, 42(2): 54-56, 68.Mei L, Zhang T, Hu H et al. Research on the monitoring and early warning mechanism of sudden respiratory infectious disease symptoms based on medical institutions[J]. Chinese Hospital Management, 2022, 42(2): 54-56, 68. -
点击查看大图
图(3) / 表(2)
计量
- 文章访问数: 49
- HTML全文浏览量: 22
- PDF下载量: 38
- 被引次数: 0