Fine-scale dengue transmission risk prediction based on multi-source geographic data in Guangzhou,China

doi:10.11853/j.issn.1003.8280.2023.05.013

Abstract

Abstract: Objective This study predicted the risk of dengue fever transmission on a fine scale, aiming to meet the practical needs of meticulous prevention and control and to provide a reference for relevant departments to formulate more precise response plans against dengue fever.Methods A Random Forest model was constructed to predict the risk of dengue fever transmission at a fine resolution of 1 km×1 km based on the data on dengue fever cases as well as natural and socio-economic factors including precipitation, land surface temperature, population density, road density, the normalized difference vegetation index, hospital accessibility, bus stop density,and the Shannon evenness index of land use in Guangzhou, China from 2017 to 2019.Results Compared with the models based on undersampling or combined sampling, the oversampling-based prediction model had better precision, with the area under the curve (AUC) being 0.999, accuracy being 0.978, precision being 0.999, recall being 0.959, and F1 value being 0.979. The analysis of the importance of single factors in dengue fever prediction revealed that the importance of population density was much higher than those of the other variables, with an average increase in mean squared error of 63.76; hospital accessibility was the second important feature;the average land surface temperature had the lowest importance among the selected variables, with an average increase in mean squared error of 35.42. The distribution of dengue fever transmission risk was consistent with the distribution of populated areas in Guangzhou. The high-risk areas accounted for 6.18% of the total area, and the at-risk populations in the high-risk areas accounted for 39.13% of the total population. More than 80.00% of the population in Yuexiu, Liwan, Haizhu, and Tianhe districts were in the high-risk areas.Conclusions The risk of dengue transmission in Guangzhou was mainly distributed in the central urban areas of Guangzhou, with Yuexiu, Liwan, and Haizhu districts as the center, extending northward to central Baiyun District, southward to the junction of Panyu and Nansha districts, and eastward to eastern Huangpu District. The predicted risk areas were highly consistent with case distributions, indicating that the method proposed in this study can more accurately depict the geographical distribution of dengue transmission risk.

Key words: Dengue fever, Transmission risk prediction, Random forest, Resampling, Guangzhou

摘要： 目的开展精细尺度登革热传播风险预测研究,为满足精细化预防和控制的现实需求和疾控部门制定更加精准的登革热疫情应对方案提供参考。方法收集广州市2017－2019年的登革热病例数据,结合降水、地表温度、人口密度、道路密度、归一化植被指数、医院可达性、公交站点密度和土地利用香农均匀度指数等自然和社会经济数据,采用随机森林模型实现1 km×1 km精细尺度登革热传播风险预测。结果基于过采样方法的预测模型精度优于欠采样和组合采样,检验受试者工作特征曲线的曲线下面积（AUC）值为0.999,准确率为0.978,精确率为0.999,查全率为0.959,F1分数值为0.979。分析单一因素对登革热预测的重要性程度发现,人口密度的重要性程度远高于其他变量,其均方误差增加量平均值为63.76。医院可达性为第2重要特征变量,平均地表温度在所选变量中重要性程度最低,其均方误差增加量平均值为35.42。广州市登革热传播风险分布与人口区分布一致,高风险区面积占总面积的6.18％,位于高风险区内的风险人口占总人口的39.13％。越秀、荔湾、海珠和天河区4个区均有80.00%以上的人口处于高风险区。结论广州市登革热传播风险区主要分布于广州市中心城区,以越秀、荔湾和海珠区为中心,向北延伸至白云区中部,向南延伸至番禺和南沙区交界处,向东延伸至黄埔区东部。预测结果中的风险区域与病例分布高度吻合,表明该研究提出的方法能够较为准确描述登革热传播风险地理分布。

关键词: 登革热, 传播风险预测, 随机森林, 重采样, 广州市

CLC Number:

R373.3⁺3

ZHANG Meng-zhen, REN Zhou-peng, FAN Jun-fu, XIAO Jian-peng, ZHANG Ying-tao. Fine-scale dengue transmission risk prediction based on multi-source geographic data in Guangzhou,China[J]. Chinese Journal of Vector Biology and Control, 2023, 34(5): 654-663.

张梦真, 任周鹏, 范俊甫, 肖建鹏, 张应涛. 基于多源地理数据的广州市精细尺度登革热传播风险预测[J]. 中国媒介生物学及控制杂志, 2023, 34(5): 654-663.

References

[1] Chen YB, Li WH, Hua JM, et al. Comparing of spatio-temporal diffusion prediction models of dengue fevers based on machine learning[J]. Geomat World, 2016, 23（6）: 8-14. DOI:10.3969/j.issn.1672-1586.2016.06.002.（in Chinese）陈业滨, 李卫红, 华家敏, 等. 基于机器学习的登革热时空扩散预测模型对比分析[J]. 地理信息世界, 2016, 23（6）: 8-14. DOI:10.3969/j.issn.1672-1586.2016.06.002.
[2] Bravo L, Roque VG, Brett J, et al. Epidemiology of dengue disease in the Philippines （2000-2011）: A systematic literature review[J]. PLoS Negl Trop Dis, 2014, 8（11）: e3027. DOI:10.1371/journal.pntd.0003027.
[3] Cheng J, Bambrick H, Yakob L, et al. Heatwaves and dengue outbreaks in Hanoi, Vietnam: New evidence on early warning[J]. PLoS Negl Trop Dis, 2020, 14（1）: e0007997. DOI:10.1371/journal.pntd.0007997.
[4] Guo CC, Zhou ZX, Wen ZH, et al. Global epidemiology of dengue outbreaks in 1990-2015: A systematic review and meta-analysis[J]. Front Cell Infect Microbiol, 2017, 7: 317. DOI:10.3389/fcimb.2017.00317.
[5] Mone FH, Hossain S, Hasan MT, et al. Sustainable actions needed to mitigate dengue outbreak in Bangladesh[J]. Lancet Infect Dis, 2019, 19（11）: 1166-1167. DOI:10.1016/S1473-3099（19）30541-9.
[6] Xu L, Stige LC, Chan KS, et al. Climate variation drives dengue dynamics[J]. Proc Natl Acad Sci USA, 2017, 114（1）: 113-118. DOI:10.1073/pnas.1618558114.
[7] Messina JP, Brady OJ, Golding N, et al. The current and future global distribution and population at risk of dengue[J]. Nat Microbiol, 2019, 4（9）: 1508-1515. DOI:10.1038/s41564-019-0476-8.
[8] Liao ZW, Wang SQ. Prevalence and prevention of major tropical diseases in China, 2000-2019[J]. China Trop Med, 2020, 20（3）: 193-201. DOI:10.13604/j.cnki.46-1064/r.2020.03.01.（in Chinese）廖志武, 王善青. 我国2000－2019年主要热带病的流行与防治概况[J]. 中国热带医学, 2020, 20（3）: 193-201. DOI:10.13604/j.cnki.46-1064/r.2020.03.01.
[9] Lai SJ, Huang ZJ, Zhou H, et al. The changing epidemiology of dengue in China, 1990-2014: A descriptive analysis of 25 years of nationwide surveillance data[J]. BMC Med, 2015, 13: 100. DOI:10.1186/s12916-015-0336-1.
[10] Zhao H, Zhang FC, Zhu Q, et al. Epidemiological and virological characterizations of the 2014 dengue outbreak in Guangzhou, China[J]. PLoS One, 2016, 11（6）: e0156548. DOI:10.1371/journal.pone.0156548.
[11] Liu QY. Dengue fever in China: New epidemical trend, challenges and strategies for prevention and control[J]. Chin J Vector Biol Control, 2020, 31（1）: 1-6. DOI:10.11853/j.issn. 1003.8280.2020.01.001.（in Chinese）刘起勇. 我国登革热流行新趋势、防控挑战及策略分析[J]. 中国媒介生物学及控制杂志, 2020, 31（1）: 1-6. DOI:10.11853/j.issn.1003.8280.2020.01.001.
[12] Shepard DS, Undurraga EA, Halasa YA, et al. The global economic burden of dengue: A systematic analysis[J]. Lancet Infect Dis, 2016, 16（8）: 935-941. DOI:10.1016/S1473-3099（16）00146-8.
[13] Morin CW, Comrie AC, Ernst K. Climate and dengue transmission: Evidence and implications[J]. Environ Health Perspect, 2013, 121（11/12）: 1264-1272. DOI:10.1289/ehp. 1306556.
[14] Mahmood S, Irshad A, Nasir JM, et al. Spatiotemporal analysis of dengue outbreaks in Samanabad town, Lahore metropolitan area, using geospatial techniques[J]. Environ Monit Assess, 2019, 191（2）: 55. DOI:10.1007/s10661-018-7162-9.
[15] Morgan J, Strode C, Salcedo-Sora JE. Climatic and socio-economic factors supporting the co-circulation of dengue, Zika and chikungunya in three different ecosystems in Colombia[J]. PLoS Negl Trop Dis, 2021, 15（3）: e0009259. DOI:10.1371/journal.pntd.0009259.
[16] Cheng J, Bambrick H, Yakob L, et al. Extreme weather conditions and dengue outbreak in Guangdong, China: Spatial heterogeneity based on climate variability[J]. Environ Res, 2021, 196: 110900. DOI:10.1016/j.envres.2021.110900.
[17] Li CX, Liu QY, Ma W. Effects of extreme precipitation events on the incidence of dengue fever in different characteristic populations in Guangzhou[J]. J Shandong Univ: Health Sci, 2021, 59（12）: 151-157. DOI:10.6040/j.issn.1671-7554.0. 2021.1013.（in Chinese）李传玺, 刘起勇, 马伟. 广州市极端降水事件对不同特征人群登革热发病的影响[J]. 山东大学学报: 医学版, 2021, 59（12）: 151-157. DOI:10.6040/j.issn.1671-7554.0.2021.1013.
[18] Akter R, Hu WB, Gatton M, et al. Climate variability, socio-ecological factors and dengue transmission in tropical Queensland, Australia: A Bayesian spatial analysis[J]. Environ Res, 2021, 195: 110285. DOI:10.1016/j.envres.2020.110285.
[19] Li CL, Wu XX, Sheridan S, et al. Interaction of climate and socio-ecological environment drives the dengue outbreak in epidemic region of China[J]. PLoS Negl Trop Dis, 2021, 15（10）: e0009761. DOI:10.1371/journal.pntd.0009761.
[20] Mudele O, Frery AC, Zanandrez LFR, et al. Dengue vector population forecasting using multisource earth observation products and recurrent neural networks[J]. IEEE J Sel Top Appl Earth Obs Remote Sens, 2021, 99: 4390-4404. DOI:10.1109/JSTARS.2021.3073351.
[21] Ogashawara I, Li L, Moreno-Madrinan MJ. Spatial-temporal assessment of environmental factors related to dengue outbreaks in São Paulo, Brazil[J]. Geohealth, 2019, 3（8）: 202-217. DOI:10.1029/2019GH000186.
[22] Pineda-Cortel MB, Clemente B, Nga PT. Modeling and predicting dengue fever cases in key regions of the Philippines using remote sensing data[J]. Asian Pac J Trop Med, 2019, 12（2）: 60-66. DOI:10.4103/1995-7645.250838.
[23] Francisco ME, Carvajal TM, Ryo M, et al. Dengue disease dynamics are modulated by the combined influences of precipitation and landscape: A machine learning approach[J]. Sci Total Environ, 2021, 792: 148406. DOI:10.1016/j.scitotenv.2021.148406.
[24] Xavier LL, Honório NA, Pessanha JFM, et al. Analysis of climate factors and dengue incidence in the metropolitan region of Rio de Janeiro, Brazil[J]. PLoS One, 2021, 16（5）: e0251403. DOI:10.1371/journal.pone.0251403.
[25] Ao LJ, Zhang YQ, Xu H, et al. Assessing the contribution of meteorology to the spatio-temporal prediction model of dengue in Guangdong province[J]. Mod Prev Med, 2020, 47（16）: 2899-2903. （in Chinese）敖琳珺, 张昱勤, 许欢, 等. 评估气象对广东省登革热时空预测模型的贡献[J]. 现代预防医学, 2020, 47（16）: 2899-2903.
[26] Watts MJ, Kotsila P, Mortyn PG, et al. Influence of socio-economic, demographic and climate factors on the regional distribution of dengue in the United States and Mexico[J]. Int J Health Geogr, 2020, 19（1）: 44. DOI:10.1186/s12942-020-00241-1.
[27] Li CL, Wu XX, Wang XF, et al. Ecological environment and socioeconomic factors drive long-term transmission and extreme outbreak of dengue fever in epidemic region of China[J]. J Cleaner Prod, 2021, 279: 123870. DOI:10.1016/j.jclepro. 2020.123870.
[28] Ren HY, Wu W, Li TG, et al. Urban villages as transfer stations for dengue fever epidemic: A case study in the Guangzhou, China[J]. PLoS Negl Trop Dis, 2019, 13（4）: e0007350. DOI:10.1371/journal.pntd.0007350.
[29] Wu PC, Lay JG, Guo HR, et al. Higher temperature and urbanization affect the spatial patterns of dengue fever transmission in subtropical Taiwan[J]. Sci Total Environ, 2009, 407（7）: 2224-2233. DOI:10.1016/j.scitotenv.2008.11.034.
[30] Wang X, Nishiura H. The epidemic risk of dengue fever in Japan: Climate change and seasonality[J]. Can J Infect Dis Med Microbiol, 2021, 2021: 6699788. DOI:10.1155/2021/6699788.
[31] Mussumeci E, Coelho FC. Large-scale multivariate forecasting models for Dengue-LSTM versus random forest regression[J]. Spat Spatiotemporal Epidemiol, 2020, 35: 100372. DOI:10. 1016/j.sste.2020.100372.
[32] Liu YQ, Liu XQ, Song WT, et al. Epidemiological analysis of dengue fever in Nanchang city, Jiangxi province from 2011 to 2019[J]. Mod Prev Med, 2021, 48（12）: 2135-2138, 2154. （in Chinese）刘仰青, 柳小青, 宋文涛, 等. 江西省南昌市2011－2019年登革热流行病学特征分析[J]. 现代预防医学, 2021, 48（12）: 2135-2138, 2154.
[33] Qi XP, Wang Y, Li Y, et al. The effects of socioeconomic and environmental factors on the incidence of dengue fever in the Pearl River Delta, China, 2013[J]. PLoS Negl Trop Dis, 2015, 9（10）: e0004159. DOI:10.1371/journal.pntd.0004159.
[34] Bouzid M, Colón-González FJ, Lung T, et al. Climate change and the emergence of vector-borne diseases in Europe: Case study of dengue fever[J]. BMC Public Health, 2014, 14: 781. DOI:10.1186/1471-2458-14-781.
[35] Chen YB, Li WH, Huang YX, et al. Spatio-temporal spreading features and the influence factors of dengue fever in downtown Guangzhou[J]. Trop Geogr, 2016, 36（5）: 767-775. DOI:10.13284/j.cnki.rddl.002881.（in Chinese）陈业滨, 李卫红, 黄玉兴, 等. 广州市登革热时空传播特征及影响因素[J]. 热带地理, 2016, 36（5）: 767-775. DOI:10.13284/j.cnki.rddl.002881.
[36] Ong J, Liu X, Rajarethinam J, et al. Mapping dengue risk in Singapore using Random Forest[J]. PLoS Negl Trop Dis, 2018, 12（6）: e0006587. DOI:10.1371/journal.pntd.0006587.
[37] Guangzhou Overview[EB/OL]. [2023-02-10]. http://www.guangzhou.gov.cn/156080.shtml.（in Chinese）中共广州市委宣传部.广州概述[EB/OL]. [2023-02-10]. http://www.guangzhou.gov.cn/156080.shtml.
[38] Peng SZ, Ding YX, Liu WZ, et al. 1 km monthly temperature and precipitation dataset for China from 1901 to 2017[J]. Earth Syst Sci Data, 2019, 11（4）: 1931-1946. DOI:10.5194/essd-11-1931-2019.
[39] USGS. MOD11A2 v006[EB/OL]. （2019-07-26） [2023-02-10]. https://lpdaac.usgs.gov/products/mod11a2v006/.
[40] Du SJ, Du SH, Liu B, et al. Large-scale urban functional zone mapping by integrating remote sensing images and open social data[J]. GISci Remote Sens, 2020, 57（3）: 411-430. DOI:10. 1080/15481603.2020.1724707.
[41] Breiman L. Random Forests[J]. Mach Learn, 2001, 45（1）: 5-32. DOI:10.1023/A:1010933404324.
[42] Ren ZP, Zhu J, Gao YF, et al. Maternal exposure to ambient PM10 during pregnancy increases the risk of congenital heart defects: Evidence from machine learning models[J]. Sci Total Environ, 2018, 630: 1-10. DOI:10.1016/j.scitotenv.2018. 02.181.
[43] Lunardon N, Menardi G, Torelli N. ROSE: A package for binary imbalanced learning[J]. R J, 2014, 6（1）: 79-89. DOI:10. 32614/RJ-2014-008.
[44] Ren ZP, Wang DQ, Ma AM, et al. Predicting malaria vector distribution under climate change scenarios in China: Challenges for malaria elimination[J]. Sci Rep, 2016, 6: 20604. DOI:10.1038/srep20604.
[45] Sang SW, Liu QY. Spatial and temporal analysis of indigenous dengue cases in Guangdong province during 2003-2012[J]. Chin J Vector Biol Control, 2015, 26（5）: 451-453. DOI:10.11853/j.issn.1003.4692.2015.05.005.（in Chinese）桑少伟, 刘起勇. 广东省2003－2012年登革热本地病例时空分析[J]. 中国媒介生物学及控制杂志, 2015, 26（5）: 451-453. DOI:10.11853/j.issn.1003.4692.2015.05.005.