Dataset Information

A dengue fever predicting model based on Baidu search index data and climate data in South China.

ABSTRACT: With the acceleration of global urbanization and climate change, dengue fever is spreading worldwide. Different levels of dengue fever have also occurred in China, especially in southern China, causing enormous economic losses. Unfortunately, there is no effective treatment for dengue, and the most popular dengue vaccine does not exhibit good curative effects. Therefore, we developed a Generalized Additive Mixed Model (GAMM) that gathered climate factors (mean temperature, relative humidity and precipitation) and Baidu search data during 2011-2015 in Guangzhou city to improve the accuracy of dengue fever prediction. Firstly, the time series dengue fever data were decomposed into seasonal, trend and remainder components by the seasonal-trend decomposition procedure based on loess (STL). Secondly, the time lag of variables was determined in cross-correlation analysis and the order of autocorrelation was estimated using autocorrelation (ACF) and partial autocorrelation functions (PACF). Finally, the GAMM was built and evaluated by comparing it with Generalized Additive Mode (GAM). Experimental results indicated that the GAMM (R2: 0.95 and RMSE: 34.1) has a superior prediction capability than GAM (R2: 0.86 and RMSE: 121.9). The study could help the government agencies and hospitals respond early to dengue fever outbreak.

SUBMITTER: Liu D

PROVIDER: S-EPMC6936853 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A dengue fever predicting model based on Baidu search index data and climate data in South China.

Liu Dan D Guo Songjing S Zou Mingjun M Chen Cong C Deng Fei F Xie Zhong Z Hu Sheng S Wu Liang L

PloS one 20191230 12

With the acceleration of global urbanization and climate change, dengue fever is spreading worldwide. Different levels of dengue fever have also occurred in China, especially in southern China, causing enormous economic losses. Unfortunately, there is no effective treatment for dengue, and the most popular dengue vaccine does not exhibit good curative effects. Therefore, we developed a Generalized Additive Mixed Model (GAMM) that gathered climate factors (mean temperature, relative humidity and ...[more]

PMID: 31887118

Similar Datasets

Project description:BackgroundFrom the beginning of 2020, the world was plunged into a pandemic caused by the novel coronavirus disease-19 (COVID-19). People increasingly searched for information related to COVID-19 on internet websites. The Baidu Index is a data sharing platform. The main data provided is the search index (SI), which represents the frequency that keywords are used in searches.MethodsJanuary 9, 2020 is an important date for the outbreak of COVID-19 in China. We compared the changes of SI before and after for 7 keywords, including "fever", "cough", "nausea", "vomiting", "abdominal pain", "diarrhea", "constipation". The slope and peak values of SI change curves are compared. Ten provinces in China were selected for a separate analysis, including Beijing, Gansu, Guangdong, Guangxi, Heilongjiang, Hubei, Sichuan, Shanghai, Xinjiang, Tibet. The change of SI was analyzed separately, and the correlation between SI and demographic and economic data was analyzed.ResultsDuring period I, from January 9 to January 25, 2020, the average daily increase (ADI) of the SI for "diarrhea" was lower than that for "cough" (889.47 vs. 1,799.12, F=11.43, P=0.002). In period II, from January 25 to April 8, 2020, the average daily decrease (ADD) of the SI for "diarrhea" was significantly lower than that for "cough", with statistical significance (cough, 191.40 vs. 441.44, F=68.66, P<0.001). The mean SI after January 9, 2020 (pre-SI) was lower than that before January 9, 2020 (post-SI) (fever, 2,616.41±116.92 vs. 3,724.51±867.81, P<0.001; cough, 3,260.04±308.43 vs. 5,590.66±874.25, P<0.001; diarrhea, 4,128.80±200.82 vs. 4,423.55±1,058.01, P<0.001). The pre-SI mean was correlated with population (P=0.004, R=0.813) and gross domestic product (GDP) (P<0.001, R=0.966). The post-SI peak was correlated with population (P=0.007, R=0.789), GDP (P=0.005, R=0.804), and previously confirmed cases (PCC) (P=0.03, R=0.670). The growth rate of the SI was correlated with the post-SI peak (P=0.04, R=0.649), PCC (P=0.003, R=0.835).ConclusionsDiarrhea was of widespread concern in all provinces before and after the COVID-19 outbreak and may be associated with novel coronavirus infection. Internet big data can reflect the public's concern about diseases, which is of great significance for the study of the epidemiological characteristics of diseases.

Project description:BackgroundLower urinary tract symptoms (LUTS) are one of the most commonly described urination disorders worldwide. Previous investigations have focused predominantly on the prospective identification of cases that meet the researchers' criteria; thus, the genuine demands regarding LUTS from patients and related issues may be neglected.ObjectiveWe aimed to examine web-based search trends and behaviors related to LUTS on a national and regional scale by using the dominant, major search engine in mainland China.MethodsBaidu Index was queried by using LUTS-related terms for the period of January 2011 to September 2020. The search volume for each term was recorded to analyze search trends and demographic distributions. For user interest, user demand graph data and trend data were collected and analyzed.ResultsOf the 13 LUTS domains, 11 domains are available in the Baidu Index database. The Baidu search index for each LUTS domain varied from 37.78% to 1.47%. The search trends for urinary frequency (2011-2018: annual percent change APC=7.82%; P<.001), incomplete emptying (2011-2014: APC=17.74%; P<.001), nocturia (2011-2018: APC=11.54%; P<.001), dysuria (2017-2020: APC=20.77%; P<.001), and incontinence (2011-2016: APC=13.39%; P<.001) exhibited fluctuations over time. The search index trends for weak stream (2011-2017: APC=-4.68%; P<.001; 2017-2020: APC=9.32%; P=.23), split stream (2011-2013: APC=9.50%; P=.44; 2013-2020: APC=2.05%; P=.71), urgency (2011-2018: APC=-2.63%; P=.03; 2018-2020: APC=8.58%; P=.19), and nocturnal enuresis (2011-2018: APC=-3.20%; P=.001; 2018-2020: APC=-4.21%; P=.04) remained relatively stable and consistent. The age distribution of the population for all LUTS-related inquiries showed that individuals aged 20 to 40 years made 73.86% (49,218,123/66,635,247) of the total search inquiries. Further, individuals aged 40 to 49 years made 12.29% (8,193,922/66,635,247) of the total search inquiries for all LUTS-related terms. People from the east part of China made 67.79% (45,172,031/66,635,247) of the total search queries. Additionally, most of the searches for LUTS-related terms were related to those for urinary diseases to varying degrees.ConclusionsWeb-based interest in LUTS-related terms fluctuated wildly and was reflected timely by Baidu Index in mainland China. The web-based search popularity of each LUTS-related term varied significantly and differed based on personal interests, the population's concerns, regional variations, and gender. These data can be used by care providers to track the prevalence of LUTS and the population's interests, guide the establishment of disease-specific health care policies, and optimize physician-patient health care sessions.

Dataset Information

A dengue fever predicting model based on Baidu search index data and climate data in South China.

Publications

A dengue fever predicting model based on Baidu search index data and climate data in South China.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets