Association between meteorological factors and the prevalence dynamics of Japanese encephalitis

Taotian Tu; Keqiang Xu; Lei Xu; Yuan Gao; Ying Zhou; Yaming He; Yang Liu; Qiyong Liu; Hengqing Ji; Wenge Tang

doi:10.1371/journal.pone.0247980

. 2021 Mar 3;16(3):e0247980. doi: 10.1371/journal.pone.0247980

Association between meteorological factors and the prevalence dynamics of Japanese encephalitis

Taotian Tu ^1,^#, Keqiang Xu ^2,^#, Lei Xu ^3,⁴, Yuan Gao ³, Ying Zhou ¹, Yaming He ¹, Yang Liu ¹, Qiyong Liu ³, Hengqing Ji ^1,^*, Wenge Tang ^1,^*

Editor: Ram K Raghavan⁵

PMCID: PMC7928514 PMID: 33657174

Abstract

Japanese encephalitis (JE) is an acute infectious disease caused by the Japanese encephalitis virus (JEV) and is transmitted by mosquitoes. Meteorological conditions are known to play a pivotal role in the spread of JEV. In this study, a zero-inflated generalised additive model and a long short-term memory model were used to assess the relationship between the meteorological factors and population density of Culex tritaeniorhynchus as well as the incidence of JE and to predict the prevalence dynamics of JE, respectively. The incidence of JE in the previous month, the mean air temperature and the average of relative humidity had positive effects on the outbreak risk and intensity. Meanwhile, the density of all mosquito species in livestock sheds (DMSL) only affected the outbreak risk. Moreover, the region-specific prediction model of JE was developed in Chongqing by used the Long Short-Term Memory Neural Network. Our study contributes to a better understanding of the JE dynamics and helps the local government establish precise prevention and control measures.

Introduction

Japanese encephalitis (JE) is an acute zoonotic disease caused by the Japanese encephalitis virus (JEV, include five serotypes) and is transmitted by mosquitoes, primarily Culex tritaeniorhynchus. It is estimated that three billion persons live in countries where the JE virus is endemic worldwide. At present, JE is prevalent mostly in 24 countries of Southeast Asia and the Western Pacific; about 68,000 clinical cases are reported annually, and the case fatality rate is 25 to 30%. About 30 to 50% of JE survivors have permanent neurological sequelae, imposing a heavy burden on public health and society [1–3]. Historically, JE was serious disease in China. The number of JE cases has declined significantly Since the implementation of the nationwide immunization program in the 1970s [4]. To date, JE is considered one of the class B notifiable infectious diseases [5]. In recent years, adults have been the population primarily affected [6]. However, JE still remains an important public health issue in China, with approximately a half of the reported cases worldwide [1].

Chongqing, China, is one of the areas with a high incidence of JE as its natural conditions, such as climate and environment, are suitable for the reproduction of mosquitoes. The incidence of JE fluctuated between 1.22 and 3.66 per 100,000 population from 1996 to 2006 in Chongqing. The vaccine for JE has been included in the Expanded Program on Immunization in Chongqing since 2007, which indicates that this vaccine has been included in the National Immunization Program for the routine vaccination of school-age children. Since then, the incidence of JE in Chongqing has decreased from 1.09 per 100,000 population in 2008 to 0.09 per 100,000 population in 2017. Although it decreased by 91.7% from 2007 to 2018, the infection rate is still the highest in Chongqing compare to other cities in China [7].

Recently, Daniel et al. found that rainfall (one-month lag), minimum temperature (six-month lag) and the Southern Oscillation Index (six-month lag) are positively associated with JE [8]. The spatial and temporal trends of the incidence of JE in the city of Chongqing are associated with temperature and rainfall [9]. Lin et al. found that the occurrence of JE is significantly associated with increasing temperature and relative humidity in Taiwan on the basis of Poisson’s regression analysis and a case-crossover study [10]. Zhang et al. showed that meteorological variables are significantly associated with the incidence of JE between 2012 and 2014 using the Bayesian conditional auto-regressive model [11]. Prediction and early warning based on multiple factors have become an interesting topic in relation to the prevention and control of mosquito-borne infectious diseases, such as malaria [12]. Previous studies have carried out to develop JE prediction models, including the compartment model and time series model. Rubel et al. showed that the compartment model presented is able to quantitatively describe the process of JE dynamics [13]. The time series model used the relationship in the sequential lag time series to predict the incidence of JE, such as Methods Autoregressive integrated moving average (ARIMA) models [14]. However, the compartment model was used for some deterministic problems and required idealized hypothetical conditions [15]. Therefore, the existing compartment model has certain defects in predicting the trend of infectious disease. In numerous cases, the time series model did not consider the relationship between infectious disease.

This study addressed the limitation by a) analyzing the correlation between meteorology, mosquito density and the incidence of JE, and b) combining JE surveillance data, meteorological data and mosquito density data with the Long-Short Term Memory (LSTM) model to predict the JE incidence accurately. The results of this study can provide scientific information that can be used by the local government to establish precise prevention and control measures.

Materials and methods

Study area

Chongqing is located in the southwest of China (Fig 1A) and in the upper reaches of the Yangtze River. It has a sub-tropical monsoon humid climate, with an annual average temperature between 16°C and 18°C and an average annual rainfall of 1,000–1,350 mm. The eastern and southern parts of Chongqing rely on two large mountains (Daba and Wuling Mountains), and the northwest and central areas consist of mainly hills and low mountains. Numerous rivers are found in the region, and the mainstream of the Yangtze River runs through the whole territory from the west to the east. Because of the diverse ecological environment and rich vegetation in Chongqing, it is conducive for the reproduction of mosquitoes.

Data collection

Human JE incidence data in Chongqing from 2007 to 2019 were available from the National Notifiable Disease Surveillance System (NNDSS). Data on mosquitoes were obtained from Chongqing Center for Disease Control and Prevention (CQCDC), including Culex tritaeniorhynchus density in human houses (CDH), Culex tritaeniorhynchus density in livestock sheds (CDL), density of all mosquito species in human houses (DMSH) and density of all mosquito species in livestock sheds (DMSL). Meteorological data were obtained from National Meteorological Information Center (NMIC) (http://data.cma.cn/site/index.html), including mean air temperature, maximum air temperature, minimum air temperature and average of relative humidity. In total, 10 valid attributes were considered in our study, and the feature parameters are shown in Table 1. We summarized the data into a monthly scale. Human JE incidence data were aggregated per month and matched to monthly surveillance data of adult mosquito population density and monthly meteorological data for the model training and prediction. All the data required by this study is available at https://doi.org/10.17026/dans-xm3-7qdm.

Table 1. The feature parameters used in this study.

Parameters	Symbol
Human JE cases per month	D
Mean air temperature	T_mean
Maximum air temperature	T_max
Minimum air temperature	T_min
Average of relative humidity	H_mean
Precipitation	P
the Culex tritaeniorhynchus density in human houses	CDH
the Culex tritaeniorhynchus density in livestock sheds	CDL
the density of all mosquito species in human houses	DMSH
the density of all mosquito species in livestock sheds	DMSL

Open in a new tab

Statistical modelling

The overall framework of this study is illustrated in Fig 2. This study analyzed the correlation between meteorology, mosquito density and the incidence of JE by using the Zero-Inflated Generalized Additive Model (ZIGAM) and constructed four candidate prediction models of JE, including LSTM model, Back Propagation Neural Network (BPNN), Gradient Boosting Machine (GBM), Generalized Additive Model (GAM) and Support Vector Regression (SVR). Based on the inverse logarithm of model outputs, the model performance and prediction accuracy were measured by RMSE.

ZIGAM is widely used in applied statistics, particularly for modelling non-linear effects of covariates in scientific and quantitative studies. Some scholars successfully demonstrated the effects of climatic conditions on the mosquito density and dengue transmission rates by used ZIGAM [16]. Data on JE were zero-inflated (Fig 1). Therefore, the incidence dynamics of JE were analyzed with ZIGAM, which consisted of a binomial and a lognormal part, which was used to explore the relationships between the incidence of JE as well as mosquito density and meteorological factors using the COZIGAM package (version 2.0.4) of R (version 3.5.1) [17]. The maximum degree of freedom for the smooth terms was set to 4 in the model selection. We used an approximation of the log of the marginal likelihood (logE) as a model selection criterion, which was correlated to the Bayesian information criterion. Models with higher logE were preferred.

LSTM is a special kind of Recurrent Neural Network (RNN). The RNN algorithm are well suited for multivariate time series data, has the ability to learn patterns and extract features from data. LSTM has been widely used in a disease diagnosis, language identification and marine temperature due to its high accuracy [18–20]. Moreover, the model can be used to predict the incidence of infectious diseases [21]. In this study, the model was used to predict the incidence of JE. When truncating the gradient where it does not cause harm, the LSTM can learn to bridge minimal time lags in excess of 1,000 discrete-time steps by enforcing constant error flow through constant error carousels within special units [22]. In order to improve the practicability of the model, we use the number of monthly cases and local environmental variables that are 12 months behind as dependent variables. The model we built can predict the trend of JE in the coming year. According to the ZIGAM study on the correlation between environmental variables and the incidence of JE, we chose the mean air temperature, average of relative humidity and DMSL as the input variables of the candidate models. The monthly data from 2007 to 2017 was used as the training set, the monthly data of 2018 was used as the verification set, and the monthly data of 2019 was used as the test set. In order to improve the prediction effect, make the model output multiple predictions by using the Dropout that can calculate the uncertainty of the LSTM forecasts and then calculate the prediction intervals.

The SVR model has shown an excellent performance for prediction data of time series. Specifically, we implemented an ε-SVR approach and tried values ranged from 0.05 to 1.0 with a span of 0.05 for the parameter C. Finally, we set the C parameter to 0.5 corresponding to the lowest RMSE value. For the BPNN model, an optimal parameter (set the number of layers to 3, the number of neurons to 6 and the learning rate to 0.02) was selected to avoid overfitting and improve the predictive performance. For the GAM model and the GBM model, the parameters of training used the default values in the python package.

Most of the experiments were run in the hardware environment with 64-bit Windows, a 3.0 GHz, Intel Core i5-8500 CPU. The BPNN model in this study were modeled through sknn library of python (version 0.7). The GBM model and SVR model in this study were modeled through scikit-learn library of python (version 0.22.2). The LSTM model in this study were modeled through TensorFlow (version 1.13.1), which is Google’s released application programming interface for deep learning.

Results

Outcomes of ZIGAM

A total of 1,531 JE cases were reported from 2007 to 2019 in Chongqing (Fig 1B). The descriptive analysis results are shown in S1 Table. The cumulative incidence (CI) increased from 0 to 2.21 per 100,000 population (Fig 1C). The monthly time series plots of all variables are shown in Fig 3. A seasonal pattern was observed among all variables.

Fig 3 — Monthly time series of the JE incidence (A), the mean air temperature (B), the maximum air temperature (C), the minimum air temperature (D), the average of relative humidity (E), the precipitation (F), the C. *tritaeniorhynchus* density in human houses (CDH) (G), the C. *tritaeniorhynchus* density in livestock sheds (CDL) (H), the density of all mosquito species in human houses (DMSH) (I) and the density of all mosquito species in livestock sheds (DMSL) (J) respectively.

Data from 2007 to 2019 were used to explore the correlation between the incidence of JE and other variables. We used the variables with no significant correlation (correlation coefficient < 0.6) in the model (S2 Table). The final ZIGAM of the JE dynamics showed that the incidence of JE in the previous month had positive effects on the outbreak risk (i.e. on the probability of incidence being > 0; Fig 4A) and outbreak intensity [i.e. on ln(incidence), given that the incidence is > 0; Fig 4E]. The mean air temperature (> 16°C) and average of relative humidity had positive effects on both outbreak risk and intensity (Fig 4B, 4C, 4F and 4G). Meanwhile, the DMSL sheds has a significant, approximately linear, and positive association with outbreak risk (Fig 4D).

Fig 4 — In order to explain the zero inflation, (A) − (D) depict a separate binomial sub-model that quantifies predictor effects on the outbreak risk (logit-scale probability of incidence > 0) and (E) − (H) depict a lognormal sub-model that quantifies predictor effects on the outbreak intensity when an outbreak occurs [ln(incidence)].

Outcomes of the LSTM model

The LSTM model can produce a prediction interval by using the dropout method in the prediction process. This method makes the model more practical. The root-mean-square error (RMSE) was utilised to evaluate the effect of prediction. The prediction results showed that the LSTM neural network models were indeed able to predict the high peaks in JE cases for these years compared to other prediction models (Fig 5 and S3 Table). In addition, Using the prediction models, we predicted the number of JE cases from January to December in 2020 (Fig 5).

Fig 5 — Data from 2008–2017 were used to train the LSTM model, data from 2018 were used as validation set and data from 2019 were utilised for model testing. In the figure, the observed values are represented by the black point, the predicted values of BPNN are represented by yellow line, the predicted values of GBM are represented by green line, the predicted values of SVR are represented by blue line, the predicted values of GAM are represented by purple line, the predicted values of LSTM are represented by red lines and the predicted interval of LSTM is represented by a light red area.

According to the model performance for the JE prediction periods of Chongqing, the LSTM model had the smallest RMSE values, which mean the discrepancy between observed value and the value predicted under the LSTM model was smallest (Fig 5 and S3 Table). The results suggested that the LSTM model outperformed other compared models and was chosen as the optimal model in this study.

Discussion

Chongqing, China, has a high incidence of JE, and a total of 1,531 cases were reported from 2007 to 2019. In this study, we not only analysed the relationship between meteorological factors and the incidence of JE using ZIGAM but also predicted the number of JE cases in Chongqing using the LSTM model. We found that the LSTM model was effective in predicting the incidence of JE.

Some studies found links between JE transmission and climate variations [23–26]. It is important to study the impact of weather on the transmission of Japanese encephalitis, as global climate change may directly or indirectly influence the mosquito density and the virus, as well as people’s behaviours [27]. In our study, the previous-month mean air temperature and previous-month average of relative humidity were found to have positive effects on the outbreak risk and intensity. The previous-month DMSL only affected the outbreak risk. This result is agreement with other studies in general [14]. Our study suggest that meteorological variables are mixed in their effect on the transmission of JE. One or two meteorological factors may play greater role in JE case occurrence or transmission than others and this finding coincides with other published study [9, 28].

Culex tritaeniorhynchus is the main vector of the JE virus in the study area, and its life cycle varies depending on the climatic. Monthly mean air temperature was found to be associated with the elevated JE incidence in Chongqing, China. The result of our study showed that the number of JE cases began to increase at temperatures was greater than 16°C. Similarly, Lin et al. found that the number of JE cases started to increase at temperatures of 22°C in Taiwan [10]. Higher temperatures, within limits, lead to more rapid development of larvae, shorter times between bloodmeals, and faster incubation times for viral infections within mosquitoes. As a result, increases in temperature allow mosquito populations to reach higher levels faster, and to be maintained for longer, thereby increasing the opportunities for viral transmission [29]. For example, for Japanese encephalitis, only 14% of mosquitoes were infected when temperatures were 18–22°C, but the figure reached 80% when temperatures were 26–30°C [30], which further verified our findings.

In our results, the average of relative humidity also had a positive impact on the transmission of JE. Relative humidity influences longevity, mating, dispersal, feeding behaviour and oviposition of mosquitoes [31]. At higher humidity, mosquitoes generally survive for longer and disperse further. Therefore, they have a greater chance of feeding on an infected animal and surviving to transmit a virus to humans or other animals. Culex tritaeniorhynchus target large animals for blood extraction, including cattle and swine, and its trajectories are mostly distributed in the wild and livestock sheds. Swine is an important reservoir host and amplification host of JEV and are the main source of JE infection. Our findings indicated that the DMSL only affected the outbreak risk in Chongqing. The positive association with DMSL was consistent with findings of other studies [32, 33].

Climatic factors can change the living and proliferation environment of mosquitoes and lead to temporal and spatial changes in mosquito density. Meteorological factors such as temperature and humidity can affect the whole life cycle of mosquitoes, accelerate or delay the growth and development of mosquito larvae, and then affect the abundance of adult mosquitoes, and further affect the abundance of next-generation mosquitoes [34, 35]. The density of the mosquito also means the probability of biting the host carrying the Japanese encephalitis virus and transmitting the virus to humans [36]. In terms of epidemiology, the total length of the development of mosquito larvae, the external incubation period of Japanese encephalitis virus in adult mosquitoes and the internal incubation period in humans is about one month [23, 37], which is consistent with the experimental results of our model.

“Deep Learning”, a branch of artificial intelligence, allows autonomously learning how viruses spread using raw observation data. Compared with the traditional statistical models, the deep learning method have many advantages, the most prominent of which is that deep learning models can automatically learn the information contained in the data without manually setting parameters such as thresholds value [22, 38]. The neural network model is established by setting the structural parameters. Adjusting the structural parameters of the model can obtain more accurate prediction results. If the established model is applied to the same task in other regions, it is only necessary to retrain the model by using the data of the other region without adjusting any parameters of model. In addition, the LSTM recursive neural network can use memory and gate units to truncate the gradient without damaging it and can bridge a large number of discrete-time steps to achieve the time series rule in a longer period of learning. Therefore, many different models have been developed by using LSTM, such as speech recognition model, forecasting model of air quality, forecasting models of infectious disease [21, 22, 38–40].

The ability to explore the relationship between climate and the incidence of infectious disease using ZIGAM has been demonstrated by previous research [17], and the LSTM neural network model has outstanding predictive power of infectious diseases compared to DNN and ARIMA [41]. In this study, we used the ZIGAM model to explore the relationship between the factors associated with JE; based on the factors with a relatively high correlation coefficient, such as mean air temperatures and mosquito density. Using the LSTM models, we assessed the number of JE cases from January 2019 to December 2020. The prediction of JE indicated that the high-risk season is in July and August. The results were consistent with that of the incidence of JE in Chongqing. Therefore, we emphasise that a set of models must be utilised to predict the number of cases in the subsequent year to improve the precision of prediction. Thus, attention should be paid and immediate actions must be taken during these months, and interventions such as health promotion and education, surveillance and control of mosquitoes and vaccination must be implemented immediately.

However, our research also had some limitations. First, the LSTM model takes a large amount of time for training compared; but the impact is not significant since the data collected in this study were from a small-sized dataset. Second, we did not consider local potential related factors such as vegetation coverage and population movements, requiring further study and improvement in the future work.

Conclusions

This study extended quantifying relationship between JE and meteorological variables based on the latest statistical analysis of multiyear time series and built a reliable prediction model of JE in Chongqing, China. The incidence of JE in the previous month, the mean air temperature and the average of relative humidity had positive effects on the outbreak risk and intensity. Meanwhile, the density of all mosquito species in livestock sheds (DMSL) only affected the outbreak risk. Moreover, we built a prediction model of JE by using LSTM in Chongqing, which enabled us to accurately predict monthly JE incidence using JE surveillance data and environmental data, including meteorology and mosquito density. According to the predicted data of model, the government can track epidemic dynamics to carry out targeted prevention and control measures. We conjecture that the result of this study could be applied more extensively in further researches.

Supporting information

S1 Table. Descriptive analysis of variables from 2007 to 2019 in Chongqing.

(DOCX)

Click here for additional data file.^{(14.2KB, docx)}

S2 Table. Correlation analysis using Pearson’s correlation test.

(DOCX)

Click here for additional data file.^{(14.8KB, docx)}

S3 Table. Comparison of model performance using the Root Mean Square Error (RMSE).

(DOCX)

Click here for additional data file.^{(17.2KB, docx)}

Acknowledgments

We would like to thank the staff at the Center for Disease Control and Prevention of Wanzhou District and Fengdu County for their support in collecting data during the study.

Data Availability

The data of this study can be found at https://doi.org/10.17026/dans-xm3-7qdm.

Funding Statement

This research was supported by the State Key Laboratory for Infectious Disease Prevention and Control Independent Fund (Contract no. 2018SKLID304). This research was partially supported by the donations from Delos Living LLC and the Cyrus Tang Foundation of Tsinghua University.

References

1.Campbell GL, Hills SL, Fischer M, Jacobson JA, Hoke CH, Hombach JM, et al. Estimated global incidence of Japanese encephalitis: A systematic review. Bull World Health Organ. 2011;89(10):766–74. 10.2471/BLT.10.085233 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Erlanger TE, Weiss S, Keiser J, Utzinger J, Wiedenmayer K. Past, present, and future of Japanese Encephalitis. Emerg Infect Dis. 2009;15(1):1–7. 10.3201/eid1501.080311 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Kumar P, Pisudde PM, Sarthi PP, Sharma MP, Keshri VR. Status and trend of acute encephalitis syndrome and Japanese encephalitis in Bihar, India. Natl Med J India. 2017;30(6):317–20. 10.4103/0970-258X.239070 [DOI] [PubMed] [Google Scholar]
4.Gao XY, Nasci R, Liang GD. The neglected arboviral infections in mainland China. PLoS Negl Trop Dis. 2010;4(4):e624. 10.1371/journal.pntd.0000624 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Zheng YY, Li MH, Wang HY, Liang GD. Japanese encephalitis and Japanese encephalitis virus in mainland China. Rev Med Virol. 2012;22(5):301–22. 10.1002/rmv.1710 [DOI] [PubMed] [Google Scholar]
6.Wang HY. The progress and challenge of Japanese encephalitis control and prevention in China. Chin J Prev Med. 2019;53(2):133–5. 10.3760/cma.j.issn.0253-9624.2019.02.002 [DOI] [PubMed] [Google Scholar]
7.Gao XY, Li XL, Li MH, Fu SH, Wang HY, Lu Z, et al. Vaccine strategies for the control and prevention of Japanese encephalitis in Mainland China, 1951–2011. PLoS Negl Trop Dis. 2014;8(8):e3015. 10.1371/journal.pntd.0003015 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Impoinvil DE, Ooi MH, Diggle PJ, Caminade C, Cardosa MJ, Morse AP, et al. The effect of vaccination coverage and climate on Japanese encephalitis in Sarawak, Malaysia. PLoS Negl Trop Dis. 2013;7(8):e2334. 10.1371/journal.pntd.0002334 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Bai YT, Xu ZG, Zhang J, Mao DQ, Luo C, He YY, et al. Regional impact of climate on Japanese encephalitis in areas located near the Three Gorges Dam. PloS ONE. 2014;9(1):e84326. 10.1371/journal.pone.0084326 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Lin CL, Chang HL, Lin CY, Chen KT. Seasonal patterns of Japanese encephalitis and associated meteorological factors in Taiwan. Int J Environ Res Public Health. 2017;14(11):1317. 10.3390/ijerph14111317 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Zhang SB, Hu WB, Qi X, Zhuang G. How socio-environmental factors are associated with Japanese encephalitis in Shaanxi, China-A Bayesian spatial analysis. Int J Environ Res Public Health. 2018;15(4):608. 10.3390/ijerph15040608 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Kumar P, Vatsa R, Sarthi PP, Kumar M, Gangare V. Modeling an association between malaria cases and climate variables for Keonjhar district of Odisha, India: A Bayesian approach. J Parasit Dis. 2020;44(2):319–31. 10.1007/s12639-020-01210-y [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Rubel F, Brugger K, Hantel M, Chvala-Mannsberger S, Bakonyi T, Weissenbock H, et al. Explaining Usutu virus dynamics in Austria: Model development and calibration. Prev Vet Med. 2008;85(3–4):166–86. 10.1016/j.prevetmed.2008.01.006 [DOI] [PubMed] [Google Scholar]
14.Lin HL, Yang LQ, Liu QY, Wang T, Hossain SR, Ho SC, et al. Time series analysis of Japanese encephalitis and weather in Linyi City, China. Int J Public Health. 2012;57(2):289–96. 10.1007/s00038-011-0236-x [DOI] [PubMed] [Google Scholar]
15.Grassly NC, Fraser C. Mathematical models of infectious disease transmission. Nat Rev Microbiol. 2008:6(6):477–87. 10.1038/nrmicro1845 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Xu L, Stige LC, Chan KS, Zhou J, Yang J, Sang SW, et al. Climate variation drives dengue dynamics. PNAS. 2017;114(1):113–8. 10.1073/pnas.1618558114 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Liu H, Chan KS. Introducing COZIGAM: An R package for unconstrained and constrained zero-inflated generalized additive model analysis. J Stat Softw. 2010;35(11):1–26.21603108 [Google Scholar]
18.Yan WZ, Zhang H, Sui J, Shen DG. Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. Med Image Comput Comput Assist Interv. 2018;11072:249–57. 10.1007/978-3-030-00931-1_29 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Zazo R, Lozano-Diez A, Gonzalez-Dominguez J, Toledano DT, Gonzalez-Rodriguez J. Language identification in short utterances using long short-term memory (LSTM) recurrent neural networks. PLoS ONE. 2016;11(1):e0146917. 10.1371/journal.pone.0146917 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Liu J, Zhang T, Han GJ, Gou Y. TD-LSTM: Temporal dependence-based LSTM networks for marine temperature prediction. Sensors. 2018;18(11):3797. 10.3390/s18113797 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Wang YB, Xu CJ, Zhang SK, Yang L, Wang ZD, Zhu Y, et al. Development and evaluation of a deep learning approach for modeling seasonality and trends in hand-foot-mouth disease incidence in mainland China. Sci Rep. 2019;9:8046. 10.1038/s41598-019-44469-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Graves A, Mohamed AR, Hinton G. Speech Recognition with Deep Recurrent Neural Networks. In: IEEE International Conference on Acoustics, Speech, and Signal Processing; 2013 May 26–31; Vancouver, Canada; 2013. p. 6645–9.
23.Flohic GL, Porphyre V, Barbazan P, Gonzalez JP. Review of climate, landscape, and viral genetics as drivers of the Japanese encephalitis virus ecology. PLoS Negl Trop Dis. 2013;7(9):e2208. 10.1371/journal.pntd.0002208 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Zhao X, Cao MQ, Feng HH, Fan H, Chen F, Feng ZJ, et al. Japanese Encephalitis risk and contextual risk factors in southwest China: A Bayesian hierarchical spatial and spatiotemporal analysis. Int J Environ Res Public Health. 2014;11(4):4201–17. 10.3390/ijerph110404201 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Wang LY, Hu WB, Magalhaes RJS, Bi P, Ding F, Sun HL, et al. The role of environmental factors in the spatial distribution of Japanese Encephalitis in mainland China. Environ Int. 2014;73:1–9. 10.1016/j.envint.2014.07.004 [DOI] [PubMed] [Google Scholar]
26.Pisudde PM, Kumar P, Sarthi PP, Deshmukh PR. Climatic determinants of Japanese encephalitis in Bihar State of India: A time-series Poisson regression analysis. J Commun Dis. 2017;49(4):13–8. [Google Scholar]
27.Murty US, Rao MS, Arunachalam N. The effects of climatic factors on the distribution and abundance of Japanese encephalitis vectors in Kurnool district of Andhra Pradesh, India. J Vector Borne Dis. 2010;47(1):26–32. [PubMed] [Google Scholar]
28.Bi P, Tong S, Donald K, Parton KA, Ni J. Climate variability and transmission of Japanese encephalitis in Eastern China. Vector-Borne Zoonotic Dis. 2003;3:111–5. 10.1089/153036603768395807 [DOI] [PubMed] [Google Scholar]
29.Russell RC. Ross River virus: ecology and distribution. Annu Rev Entomol. 2002;47:1–31. 10.1146/annurev.ento.47.091201.145100 [DOI] [PubMed] [Google Scholar]
30.Chen BQ. Epidemiology. 2nd ed. Geng GY, editor. Beijing: People’s Medical Publishing House; 1996. [Google Scholar]
31.McMichael AJ, Haines A, Kovats RS, Slooff R. Climate changes and human health. Geneva: WHO; 1996. [Google Scholar]
32.Borah J, Dutta P, Khan SA, Mahanta J. Association of weather and anthropogenic factors for transmission of Japanese encephalitis in an endemic area of India. EcoHealth. 2013;10(2):129–36. 10.1007/s10393-013-0849-z [DOI] [PubMed] [Google Scholar]
33.Ren XJ, Fu SH, Dai PF, Wang HY, Li YY, Li XL, et al. Pigsties near dwellings as a potential risk factor for the prevalence of Japanese encephalitis virus in adult in Shanxi, China. Infect Dis Poverty. 2017;6(1):100. 10.1186/s40249-017-0312-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Marini G, Manica M, Arnoldi D, Inama E, Rosà R, Rizzoli A. Influence of temperature on the life-cycle dynamics of Aedes albopictus population established at temperate latitudes: A laboratory experiment. Insects. 2020;11(11):808. 10.3390/insects11110808 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Roche B, Léger L, L’Ambert G, Lacour G, Foussadier R, Besnard G, et al. The spread of Aedes albopictus in metropolitan France: Contribution of environmental drivers and human activities and predictions for a near future. PLoS ONE. 2015;10(5):e0125600. 10.1371/journal.pone.0125600 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Tuno N, Tsuda Y, Takagi M. How zoophilic Japanese encephalitis vector mosquitoes feed on humans. J Med Entomol. 2017;54(1):8–13. 10.1093/jme/tjw165 [DOI] [PubMed] [Google Scholar]
37.Eckhoff PA. A malaria transmission-directed model of mosquito life cycle and ecology. Malaria Journal. 2011;10:303. 10.1186/1475-2875-10-303 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Xu JC, Xu KQ, Li ZC, Meng FX, Tu TT, Xu L, et al. Forecast of dengue cases in 20 Chinese cities based on the deep learning method. Int J Environ Res Public Health. 2020;17(2):453. 10.3390/ijerph17020453 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Jordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science. 2015;349(6245):255–60. 10.1126/science.aaa8415 [DOI] [PubMed] [Google Scholar]
40.Freeman BS, Taylor G, Gharabaghi B, Thé J. Forecasting air quality time series using deep learning. J Air Waste Manag Assoc. 2018;68(8):866–86. 10.1080/10962247.2018.1459956 [DOI] [PubMed] [Google Scholar]
41.Chae S, Kwon S, Lee D. Predicting infectious disease using deep learning and big data. Int J Environ Res Public Health. 2018;15(8):1596. 10.3390/ijerph15081596 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0247980.r001

Decision Letter 0

Ram K Raghavan

24 Aug 2020

PONE-D-20-17377

Association between climate factors and the prevalence dynamics of Japanese Encephalitis

PLOS ONE

Dear Dr. Tu,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 08 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Ram K. Raghavan

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

2.1. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2.2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

"This research was supported by the State Key Laboratory for Infectious Disease

Prevention and Control Independent Fund (Contract no. 2018SKLID304). This

research was partially supported by the donations from Delos Living LLC and the

Cyrus Tang Foundation of Tsinghua University."

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"YES - Specify the role(s) played."

Additionally, because some of your funding information pertains to commercial funding, we ask you to provide an updated Competing Interests statement, declaring all sources of commercial funding.

In your Competing Interests statement, please confirm that your commercial funding does not alter your adherence to PLOS ONE Editorial policies and criteria by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests. If this statement is not true and your adherence to PLOS policies on sharing data and materials is altered, please explain how.

Please include the updated Competing Interests Statement and Funding Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

6. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript titled Association between climate factors and the prevalence dynamics of Japanese Encephalitis incidences (in terms of mosquito density) by Tu et al. presents an approach of nonlinear statistical modeling in predicting the JE dynamics over the region Chongqing (southwest part in China). Author considers the data for the period of 13 years for JE incidences and meteorological par ammeters.

However, the article is written very simply, and I observe not up to the standard of the journal. There are some grammatical issues along lacking the proper definition of study gap.

1. The term climate factor should be replaced with ‘meteorological factor’.

2. The author applies GAM to study the relationship for vector density, and long short term memory model (LSTM) for JE incidence with met. Parameters. No justification for adopting two different modeling aspect.

3. The predictive ability of the model should be written abstract line, and author avoid to do this.

4. The Introduction section is not structured. Globally published literature is missing. The hierarchy should be, global aspect, then continental, then country, then province/district. The author fails to perform this hierarchy. Some

5. Authors wrote, “However, no studies on the prediction of the incidence of JE, which is based on 84 factors such as mosquito density and climate, have been carried out” (pg. 4, line 83-84). In what context the author states this sentence, global or local. Around the globe, researchers have already contributed the stochastic modeling skill for association of JE cases and meteorological parameters (Lin et al., 2012; Pisudde et al., 2017). In what respect the author would like to state that. Justify and clarify.

6. Line #83-88. Is there only the meteorological factors that affect JE vector density for JE incidences? The several other factors like land use land cover changes, built dam, water bodies, socioeconomic status of the exposed population, intervention schemes/policy. Justify, why authors do not take into account, all these?

7. Why author opt for stochastic approach over the deterministic approach, the compartment modeling (deterministic) is also another way to model perfectly. Justification needed.

8. Line #89, instead of Methods, there should be Study Area, Data and Methods. The further subsection may be there.

9. The temporal resolution of data is missing in the ‘Data Collection’ section. For the ease of the scientific community, this should be there. A table may be presented here, for data sources of JE incidences as well as meteorological variables, along with the temporal resolution (daily/weekly/fortnightly/monthly/yearly).

10. The models’ outcome in the result section is not expressed much. The author was only direct to see the tables and figures. The crucial observations are not expressed.

11. The result and methods are very inconsistent. The result outcomes are self-contradictory. The result need to restructure.

12. Discussion is poorly written. Author try to make this very lengthy instead of technical discussion and suggestion. The scientific approach is missing in the discussion part.

13. The conclusion section is missing.

Lin, H., Yang, L., Liu, Q., Wang, T., Hossain, S.R., Ho, S.C., Tian, L., 2012. Time series analysis of Japanese encephalitis and weather in Linyi City, China. Int. J. Public Health 57, 289–296. https://doi.org/10.1007/s00038-011-0236-x

Pisudde, P.M., Kumar, P., Sarthi, P.P., Deshmukh, P.R., 2017. Climatic Determinants of Japanese Encephalitis in Bihar State of India : A Time-Series Poisson Regression Analysis. J. Commun. Dis. 49, 13–18. https://doi.org/10.24321/0019.5138.201729

Reviewer #2: The manuscript presents the results of a correlative analysis between selected climate variables and case incidence of Japanese Encephalitis in central China. The paper addresses a topic that is appropriate for the journal and potentially supplies important information for both basic scientific understanding of environmental influences on disease and for potential application to controlling the disease in this area. The paper compares several methods for deriving correlative models between climate and JE incidence, and does in fact provide the reader with an evaluation of the relative effectiveness of these models. However, the paper paper is frustratingly vague and difficult to follow at times, particularly in its coverage of the basic methodology and in its summary of how the climate variables are related to disease incidence.

The major issue with the paper is that it’s just very difficult to follow exactly what the authors did. Figure 2 shows a flow chart of the methods, showing the sources of data and the general flow of the analysis. It would be helpful to include a table in the body of the paper (rather than the supporting materials) that shows exactly w hat meteorological data are included. There are hints in the paper (temperature extrema seem to be one of the variables) but the reader never really knows for sure, and this becomes a problem later in trying to suss out what the findings of the study really are. All of the data are apparently funneled through the ZIGAM method, but the methods section of the paper never really explains what this is or what it does. It is defined on lines 79-80, and some references are cited. Additionally, in lines 130-141 some additional information is provided. Unfortunately, as a person wholly unfamiliar with this technique, this information was of little use. This may not be the case for readers more familiar with this type of analysis, but for those who are not knowledgeable in this area of analysis, the entire methodology description is not very helpful, at best. Since the entire analysis funnels though this technique, I would strongly urge the authors to write a more complete explanation of what this method does, and why it was chosen for this analysis. Otherwise, the reader is left with the frustrating task of searching through several more papers in order to understand what is in this one.

I had similar issues with the four methods for developing predictive models, LSTM, BPNN, GBM, and SVR. The way the paper is written seems to suggest the LSTM is somehow considered a standard in this type of analysis, and that the other three methods are presented as potentially superior alternatives? Is this correct? IF so, I would suggest that this be made clearer to the reader. On page 7, the paragraph beginning on line 142, some descriptive material for LSTM is provided, but again, in such a way as to be of little help to an unfamiliar reader. I would also note that there is a lot of detail about the other three methods (BPNN, GBM, and SVR) that is simply glossed over. I have only limited experience with neural networks and gradient boosting, but have worked rather extensively with support vector regression, enough to understand that parameter tuning and kernal basis functions are crucial to understanding the effectiveness of them. Absolutely no indication of any of this information is provided. At minimum, one should report the kernal used and the gamma and r parameters. Ideally, the method for determining these parameters should be provided. I suspect that similar information should be provided about the other two methods (GBM, BPNN). Without this type of information, the analysis cannot be replicated, nor does omitting them help other researcher who may wish to use this approach for similar analyses.

In addition to the lack of relevant detail in the methods section, the actual results also lack some specificity. The purpose of the paper was to model the relationship between climate and JE incidence, and to predict the occurrence of cases. Judging from Figure 5, the latter goal was achieved. Unfortunately, it is not really clear how it was achieved. Which climate variables were most relevant, and how relevant were they? Hard to really tell from the body of the manuscript. In the abstract (lines 27-30) it is noted that mean temperature and humidity are important (as is density of mosquito species in livestock sheds, a variable whose provenance is not very clear). In the body of the manuscript, though, this information is very difficult to separate out. Clearly, the authors have developed a method that is potentially useful for predicting JE in this study area, and potentially useful for researchers who may want to understand this disease (and possibly other, similar, mosquito-borne diseases) in other places. Without much clearer elucidation of their methods, this would be difficult to do. One of my “acid tests” for reading a manuscript is, “given similar data, could I repeat their methods, even if I had never used them before.” In this case the answer is no. Even if I were to track down all of the literature cited to support the methods, It’s still not clear how the authors reached their conclusions.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Dr. Praveen Kumar

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Mar 3;16(3):e0247980. doi: 10.1371/journal.pone.0247980.r002

Author response to Decision Letter 0

3 Nov 2020

November 3, 2020

PLoS ONE

Dear Editors,

On behalf of my co-authors, we thank you very much for giving us an opportunity to revise our manuscript, we appreciate the editor for their meticulous working attitude.

We redrew Figure 1 using the ArcGIS (version 10.6). Therefore, the figure has no copyright issues.

I look forward to hearing from you again.

Best regards,

Taotian Tu (taotiantu@sina.cn)

Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China;

Wenge Tang (twg@cqcdc.org)

Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China.

Response to Reviewer #1:

We would like to thank the reviewer for careful and thorough reading of this manuscript and for the thoughtful comments and constructive suggestions, which help to improve the quality of this manuscript.

At the following, the points mentioned by the reviewers will be discussed:

1. The term climate factor should be replaced with ‘meteorological factor’.

[Reply] Thank you for your advice. After careful consideration, we also think it is more reasonable to change " climate factor " to " meteorological factor ". We have made changes in main text (line1, 23, 26, 82, 237, 301).

[Reply] Thank you for your advice. This study analyzed the correlation between meteorological variables and the incidence of JE by using the ZIGAM. With the help of the research results, we use LSTM to predict the occurrence of JE in the next year, so that the prediction will be more accurate. We added a comparison of the LSTM and GAM models in S3 table and Figure 5.

3. The predictive ability of the model should be written abstract line, and author avoid to do this.

[Reply] Thank you for your advice. We have made changes in the figure of the predictive ability of the model (Figure 5).

[Reply] Thank you for your advice. We have made changes according to your advice (line 37-49).

5. Authors wrote, “However, no studies on the prediction of the incidence of JE, which is based on 84 factors such as mosquito density and climate, have been carried out” (pg. 4. line (83-84). In what context the author states this sentence, global or local. Around the globe, researchers have already contributed the stochastic modeling skill for association of JE cases and meteorological parameters (Lin et al., 2012; Pisudde et al., 2017). In what respect the author would like to state that. Justify and clarify.

[Reply] Thank you for your advice. We have made corresponding amendments according to your advice (line 70-79).

[Reply] Thank you for your advice. Due to the problem of data acquisition, our research did not consider these factors that may have a certain impact on the occurrence of JE. We also mentioned these factors in the discussion section, requiring further study and improvement in the future work (line 295-299).

7. Why author opt for stochastic approach over the deterministic approach, the compartment modeling (deterministic) is also another way to model perfectly. Justification needed.

[Reply] Thank you for your advice. We observed that the compartment model was used for some deterministic problems and required idealized hypothetical conditions. Therefore, the existing compartment model has certain defects in predicting the trend of infectious disease (line 75-79).

8. Line #89, instead of Methods, there should be Study Area, Data and Methods. The further subsection may be there.

[Reply] Thank you for your advice. We have made corresponding amendments according to your advice (line 86).

[Reply] Thank you for your advice. We have added the corresponding content according to your advice (line 113-117).

10. The models’ outcome in the result section is not expressed much. The author was only direct to see the tables and figures. The crucial observations are not expressed.

[Reply] Thank you for your advice. We have added the corresponding content according to your advice (line 195-204 and line 229-233).

11.The result and methods are very inconsistent. The result outcomes are self-contradictory. The result need to restructure.

[Reply] Thank you for your advice. We have modified the content of the results (line 195-204 and line 229-233).

12. Discussion is poorly written. Author try to make this very lengthy instead of technical discussion and suggestion. The scientific approach is missing in the discussion part.

[Reply] Thank you for your advice. We have made major revisions to the discussion (line 235-299).

13. The conclusion section is missing.

[Reply] Thank you for your advice. We have added the conclusions (line 301-310).

Response to Reviewer #2:

At the following, the points mentioned by the reviewers will be discussed:

1. The manuscript presents the results of a correlative analysis between selected climate variables and case incidence of Japanese Encephalitis in central China. The paper addresses a topic that is appropriate for the journal and potentially supplies important information for both basic scientific understanding of environmental influences on disease and for potential application to controlling the disease in this area. The paper compares several methods for deriving correlative models between climate and JE incidence, and does in fact provide the reader with an evaluation of the relative effectiveness of these models. However, the paper is frustratingly vague and difficult to follow at times, particularly in its coverage of the basic methodology and in its summary of how the climate variables are related to disease incidence.

[Reply] Thank you for your advice. We have introduced the research methods more clearly (line 121-180), and elaborated on the relationship between meteorological factors, mosquitoes and Japanese encephalitis in the discussion section of the manuscript (line 240-272).

2. The major issue with the paper is that it’s just very difficult to follow exactly what the authors did. Figure 2 shows a flow chart of the methods, showing the sources of data and the general flow of the analysis. It would be helpful to include a table in the body of the paper (rather than the supporting materials) that shows exactly w hat meteorological data are included. There are hints in the paper (temperature extrema seem to be one of the variables) but the reader never really knows for sure, and this becomes a problem later in trying to suss out what the findings of the study really are. All of the data are apparently funneled through the ZIGAM method, but the methods section of the paper never really explains what this is or what it does. It is defined on lines 79-80, and some references are cited. Additionally, in lines 130-141 some additional information is provided. Unfortunately, as a person wholly unfamiliar with this technique, this information was of little use. This may not be the case for readers more familiar with this type of analysis, but for those who are not knowledgeable in this area of analysis, the entire methodology description is not very helpful, at best. Since the entire analysis funnels though this technique, I would strongly urge the authors to write a more complete explanation of what this method does, and why it was chosen for this analysis. Otherwise, the reader is left with the frustrating task of searching through several more papers in order to understand what is in this one.

[Reply] Thank you for careful reading of the manuscript. We have made a table in the body of the manuscript, which can accurately display the included meteorological variables (Table 1, line118). Meanwhile, we have further explained the methods of statistical modeling, including ZIGAM , LSTM and other model (line121-180).

3. I had similar issues with the four methods for developing predictive models, LSTM, BPNN, GBM, and SVR. The way the paper is written seems to suggest the LSTM is somehow considered a standard in this type of analysis, and that the other three methods are presented as potentially superior alternatives? Is this correct? IF so, I would suggest that this be made clearer to the reader. On page 7, the paragraph beginning on line 142, some descriptive material for LSTM is provided, but again, in such a way as to be of little help to an unfamiliar reader. I would also note that there is a lot of detail about the other three methods (BPNN, GBM, and SVR) that is simply glossed over. I have only limited experience with neural networks and gradient boosting, but have worked rather extensively with support vector regression, enough to understand that parameter tuning and kernal basis functions are crucial to understanding the effectiveness of them. Absolutely no indication of any of this information is provided. At minimum, one should report the kernal used and the gamma and r parameters. Ideally, the method for determining these parameters should be provided. I suspect that similar information should be provided about the other two methods (GBM, BPNN). Without this type of information, the analysis cannot be replicated, nor does omitting them help other researcher who may wish to use this approach for similar analyses.

[Reply] Thank you for your advice. We have added the detailed explanation of the parameter setting of SVR, BPNN, GBM and GAM (line 167-180).

4. In addition to the lack of relevant detail in the methods section, the actual results also lack some specificity. The purpose of the paper was to model the relationship between climate and JE incidence, and to predict the occurrence of cases. Judging from Figure 5, the latter goal was achieved. Unfortunately, it is not really clear how it was achieved. Which climate variables were most relevant, and how relevant were they? Hard to really tell from the body of the manuscript. In the abstract (lines 27-30) it is noted that mean temperature and humidity are important (as is density of mosquito species in livestock sheds, a variable whose provenance is not very clear). In the body of the manuscript, though, this information is very difficult to separate out. Clearly, the authors have developed a method that is potentially useful for predicting JE in this study area, and potentially useful for researchers who may want to understand this disease (and possibly other, similar, mosquito-borne diseases) in other places. Without much clearer elucidation of their methods, this would be difficult to do. One of my “acid tests” for reading a manuscript is, “given similar data, could I repeat their methods, even if I had never used them before.” In this case the answer is no. Even if I were to track down all of the literature cited to support the methods, It’s still not clear how the authors reached their conclusions.

[Reply] Thank you for your advice. We have supplemented the content in statistical modelling (lines 121-180). In addition, we have agreed to upload a minimal set of anonymous data. The data of this study can be found at https://doi.org/10.17026/dans-xm3-7qdm

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(20.9KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0247980.r003

Decision Letter 1

Ram K Raghavan

11 Jan 2021

PONE-D-20-17377R1

Association between meteorological factors and the prevalence dynamics of Japanese Encephalitis

PLOS ONE

Dear Dr. Tu,

Reviewer 1 has raised some minor questions that I think can be addressed very easily.

First of Reviewer 1's comment suggests that you cite a couple of articles. Please know (and it is my recommendation) that you DO NOT have to cite the two articles as these papers do not directly inform your work. As far the other comments, please prepare a response and make any changes to the manuscript if you collectively deem necessary.

We appreciate your patience with this review process.

Please submit your revised manuscript by Feb 25 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Ram K. Raghavan

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: No

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: 1. The literature survey still missing the several existing research that have implemented statistical modeling for the association with other Vector Borne Disease like malaria. Author may thoroughly revisit the introduction section and should add some more recent research that focus on Biometeorology. I am suggesting some of the recent findings.

a. Kumar, P., Vatsa, R., Sarthi, P. P., Kumar, M., & Gangare, V. (2020). Modeling an association between malaria cases and climate variables for Keonjhar district of Odisha, India: a Bayesian approach. Journal of Parasitic Diseases. https://doi.org/10.1007/s12639-020-01210-y

b. Kumar, P., Pisudde, P. M., Sarthi, P. P., Sharma, M. P., & Keshri, V. R. (2017). Acute encephalitis syndrome and Japanese Encephalitis, status and trends in Bihar State, India. THE NATIONAL MEDICAL JOURNAL OF INDIA, 30, 317–320. https://doi.org/10.4103/0970-258X.239070

2. How do the meteorological variable play role in the life cycle of JE virus Vector, author should must address here.

3. Do different geography have different threshold for the rainfall, temperature, relative humidity, etc. Author should must discuss in the discussion section.

4. How the build model may be operationalized for public services? Is there any such scope? Brief in short, the application of your model in real time.

5. Author say there is a positive effect on the outbreak with a lag 1-month of some meteorological variable. Is this really feasible, as the mosquito complete the life cycle within a month. A justification should be added.

I would like to see the further changes done by the author.

Reviewer #2: My previous concerns with the paper were that it lacked important details on the data used and the analysis methods. These have now been addressed. There is no further revision needed.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Praveen Kumar

Reviewer #2: No

PLoS One. 2021 Mar 3;16(3):e0247980. doi: 10.1371/journal.pone.0247980.r004

Author response to Decision Letter 1

14 Feb 2021

February 9, 2021

PLoS ONE

Dear Editors,

On behalf of my co-authors, we thank you very much for giving us an opportunity to revise our manuscript, we appreciate the editor for their meticulous working attitude.

We have replied the point-by-point to the reviewers' comments and revised the manuscript carefully. the comments of all reviewers are all valuable and very helpful for revising and improving our paper.

We hope that the revised manuscript could meet the high standard of your prestigious journal.

Best regards,

Taotian Tu (taotiantu@sina.cn)

Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China;

Wenge Tang (twg@cqcdc.org)

Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China.

Response to Reviewer #1:

At the following, the points mentioned by the reviewers will be discussed:

1.The literature survey still missing the several existing research that have implemented statistical modeling for the association with other Vector Borne Disease like malaria. Author may thoroughly revisit the introduction section and should add some more recent research that focus on Biometeorology. I am suggesting some of the recent findings.

[Reply] Thank you for your advice. We have added most recent publications on Biometeorology in the introduction (line 44, 70, line 345-347 and line 369-371).

2. How do the meteorological variable play role in the life cycle of JE virus Vector, author should must address here.

[Reply] Thank you for your advice. We have added the corresponding content according to your advice (line 252-253, line 276-286).

3. Do different geography have different threshold for the rainfall, temperature, relative humidity, etc. Author should must discuss in the discussion section.

[Reply] Thank you for your advice. We have revised the corresponding content of the discussion (line 289-296).

4. How the build model may be operationalized for public services? Is there any such scope? Brief in short, the application of your model in real time.

[Reply] Thank you for your advice. We have made changes according to your advice (line 299-301, line 330-332).

5.Author say there is a positive effect on the outbreak with a lag 1-month of some meteorological variable. Is this really feasible, as the mosquito complete the life cycle within a month. A justification should be added.

[Reply] Thank you for your advice. We have added the corresponding content according to your advice (line 276-286).

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(19.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0247980.r005

Decision Letter 2

Ram K Raghavan

18 Feb 2021

Association between meteorological factors and the prevalence dynamics of Japanese Encephalitis

PONE-D-20-17377R2

Dear Dr. Tu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ram K. Raghavan

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0247980.r006

Acceptance letter

Ram K Raghavan

22 Feb 2021

PONE-D-20-17377R2

Association between meteorological factors and the prevalence dynamics of Japanese Encephalitis

Dear Dr. Tu:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ram K. Raghavan

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Descriptive analysis of variables from 2007 to 2019 in Chongqing.

(DOCX)

Click here for additional data file.^{(14.2KB, docx)}

S2 Table. Correlation analysis using Pearson’s correlation test.

(DOCX)

Click here for additional data file.^{(14.8KB, docx)}

S3 Table. Comparison of model performance using the Root Mean Square Error (RMSE).

(DOCX)

Click here for additional data file.^{(17.2KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(20.9KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(19.2KB, docx)}

Data Availability Statement

The data of this study can be found at https://doi.org/10.17026/dans-xm3-7qdm.

[pone.0247980.ref001] 1.Campbell GL, Hills SL, Fischer M, Jacobson JA, Hoke CH, Hombach JM, et al. Estimated global incidence of Japanese encephalitis: A systematic review. Bull World Health Organ. 2011;89(10):766–74. 10.2471/BLT.10.085233 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref002] 2.Erlanger TE, Weiss S, Keiser J, Utzinger J, Wiedenmayer K. Past, present, and future of Japanese Encephalitis. Emerg Infect Dis. 2009;15(1):1–7. 10.3201/eid1501.080311 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref003] 3.Kumar P, Pisudde PM, Sarthi PP, Sharma MP, Keshri VR. Status and trend of acute encephalitis syndrome and Japanese encephalitis in Bihar, India. Natl Med J India. 2017;30(6):317–20. 10.4103/0970-258X.239070 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref004] 4.Gao XY, Nasci R, Liang GD. The neglected arboviral infections in mainland China. PLoS Negl Trop Dis. 2010;4(4):e624. 10.1371/journal.pntd.0000624 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref005] 5.Zheng YY, Li MH, Wang HY, Liang GD. Japanese encephalitis and Japanese encephalitis virus in mainland China. Rev Med Virol. 2012;22(5):301–22. 10.1002/rmv.1710 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref006] 6.Wang HY. The progress and challenge of Japanese encephalitis control and prevention in China. Chin J Prev Med. 2019;53(2):133–5. 10.3760/cma.j.issn.0253-9624.2019.02.002 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref007] 7.Gao XY, Li XL, Li MH, Fu SH, Wang HY, Lu Z, et al. Vaccine strategies for the control and prevention of Japanese encephalitis in Mainland China, 1951–2011. PLoS Negl Trop Dis. 2014;8(8):e3015. 10.1371/journal.pntd.0003015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref008] 8.Impoinvil DE, Ooi MH, Diggle PJ, Caminade C, Cardosa MJ, Morse AP, et al. The effect of vaccination coverage and climate on Japanese encephalitis in Sarawak, Malaysia. PLoS Negl Trop Dis. 2013;7(8):e2334. 10.1371/journal.pntd.0002334 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref009] 9.Bai YT, Xu ZG, Zhang J, Mao DQ, Luo C, He YY, et al. Regional impact of climate on Japanese encephalitis in areas located near the Three Gorges Dam. PloS ONE. 2014;9(1):e84326. 10.1371/journal.pone.0084326 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref010] 10.Lin CL, Chang HL, Lin CY, Chen KT. Seasonal patterns of Japanese encephalitis and associated meteorological factors in Taiwan. Int J Environ Res Public Health. 2017;14(11):1317. 10.3390/ijerph14111317 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref011] 11.Zhang SB, Hu WB, Qi X, Zhuang G. How socio-environmental factors are associated with Japanese encephalitis in Shaanxi, China-A Bayesian spatial analysis. Int J Environ Res Public Health. 2018;15(4):608. 10.3390/ijerph15040608 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref012] 12.Kumar P, Vatsa R, Sarthi PP, Kumar M, Gangare V. Modeling an association between malaria cases and climate variables for Keonjhar district of Odisha, India: A Bayesian approach. J Parasit Dis. 2020;44(2):319–31. 10.1007/s12639-020-01210-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref013] 13.Rubel F, Brugger K, Hantel M, Chvala-Mannsberger S, Bakonyi T, Weissenbock H, et al. Explaining Usutu virus dynamics in Austria: Model development and calibration. Prev Vet Med. 2008;85(3–4):166–86. 10.1016/j.prevetmed.2008.01.006 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref014] 14.Lin HL, Yang LQ, Liu QY, Wang T, Hossain SR, Ho SC, et al. Time series analysis of Japanese encephalitis and weather in Linyi City, China. Int J Public Health. 2012;57(2):289–96. 10.1007/s00038-011-0236-x [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref015] 15.Grassly NC, Fraser C. Mathematical models of infectious disease transmission. Nat Rev Microbiol. 2008:6(6):477–87. 10.1038/nrmicro1845 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref016] 16.Xu L, Stige LC, Chan KS, Zhou J, Yang J, Sang SW, et al. Climate variation drives dengue dynamics. PNAS. 2017;114(1):113–8. 10.1073/pnas.1618558114 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref017] 17.Liu H, Chan KS. Introducing COZIGAM: An R package for unconstrained and constrained zero-inflated generalized additive model analysis. J Stat Softw. 2010;35(11):1–26.21603108 [Google Scholar]

[pone.0247980.ref018] 18.Yan WZ, Zhang H, Sui J, Shen DG. Deep chronnectome learning via full bidirectional long short-term memory networks for MCI diagnosis. Med Image Comput Comput Assist Interv. 2018;11072:249–57. 10.1007/978-3-030-00931-1_29 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref019] 19.Zazo R, Lozano-Diez A, Gonzalez-Dominguez J, Toledano DT, Gonzalez-Rodriguez J. Language identification in short utterances using long short-term memory (LSTM) recurrent neural networks. PLoS ONE. 2016;11(1):e0146917. 10.1371/journal.pone.0146917 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref020] 20.Liu J, Zhang T, Han GJ, Gou Y. TD-LSTM: Temporal dependence-based LSTM networks for marine temperature prediction. Sensors. 2018;18(11):3797. 10.3390/s18113797 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref021] 21.Wang YB, Xu CJ, Zhang SK, Yang L, Wang ZD, Zhu Y, et al. Development and evaluation of a deep learning approach for modeling seasonality and trends in hand-foot-mouth disease incidence in mainland China. Sci Rep. 2019;9:8046. 10.1038/s41598-019-44469-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref022] 22.Graves A, Mohamed AR, Hinton G. Speech Recognition with Deep Recurrent Neural Networks. In: IEEE International Conference on Acoustics, Speech, and Signal Processing; 2013 May 26–31; Vancouver, Canada; 2013. p. 6645–9.

[pone.0247980.ref023] 23.Flohic GL, Porphyre V, Barbazan P, Gonzalez JP. Review of climate, landscape, and viral genetics as drivers of the Japanese encephalitis virus ecology. PLoS Negl Trop Dis. 2013;7(9):e2208. 10.1371/journal.pntd.0002208 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref024] 24.Zhao X, Cao MQ, Feng HH, Fan H, Chen F, Feng ZJ, et al. Japanese Encephalitis risk and contextual risk factors in southwest China: A Bayesian hierarchical spatial and spatiotemporal analysis. Int J Environ Res Public Health. 2014;11(4):4201–17. 10.3390/ijerph110404201 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref025] 25.Wang LY, Hu WB, Magalhaes RJS, Bi P, Ding F, Sun HL, et al. The role of environmental factors in the spatial distribution of Japanese Encephalitis in mainland China. Environ Int. 2014;73:1–9. 10.1016/j.envint.2014.07.004 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref026] 26.Pisudde PM, Kumar P, Sarthi PP, Deshmukh PR. Climatic determinants of Japanese encephalitis in Bihar State of India: A time-series Poisson regression analysis. J Commun Dis. 2017;49(4):13–8. [Google Scholar]

[pone.0247980.ref027] 27.Murty US, Rao MS, Arunachalam N. The effects of climatic factors on the distribution and abundance of Japanese encephalitis vectors in Kurnool district of Andhra Pradesh, India. J Vector Borne Dis. 2010;47(1):26–32. [PubMed] [Google Scholar]

[pone.0247980.ref028] 28.Bi P, Tong S, Donald K, Parton KA, Ni J. Climate variability and transmission of Japanese encephalitis in Eastern China. Vector-Borne Zoonotic Dis. 2003;3:111–5. 10.1089/153036603768395807 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref029] 29.Russell RC. Ross River virus: ecology and distribution. Annu Rev Entomol. 2002;47:1–31. 10.1146/annurev.ento.47.091201.145100 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref030] 30.Chen BQ. Epidemiology. 2nd ed. Geng GY, editor. Beijing: People’s Medical Publishing House; 1996. [Google Scholar]

[pone.0247980.ref031] 31.McMichael AJ, Haines A, Kovats RS, Slooff R. Climate changes and human health. Geneva: WHO; 1996. [Google Scholar]

[pone.0247980.ref032] 32.Borah J, Dutta P, Khan SA, Mahanta J. Association of weather and anthropogenic factors for transmission of Japanese encephalitis in an endemic area of India. EcoHealth. 2013;10(2):129–36. 10.1007/s10393-013-0849-z [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref033] 33.Ren XJ, Fu SH, Dai PF, Wang HY, Li YY, Li XL, et al. Pigsties near dwellings as a potential risk factor for the prevalence of Japanese encephalitis virus in adult in Shanxi, China. Infect Dis Poverty. 2017;6(1):100. 10.1186/s40249-017-0312-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref034] 34.Marini G, Manica M, Arnoldi D, Inama E, Rosà R, Rizzoli A. Influence of temperature on the life-cycle dynamics of Aedes albopictus population established at temperate latitudes: A laboratory experiment. Insects. 2020;11(11):808. 10.3390/insects11110808 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref035] 35.Roche B, Léger L, L’Ambert G, Lacour G, Foussadier R, Besnard G, et al. The spread of Aedes albopictus in metropolitan France: Contribution of environmental drivers and human activities and predictions for a near future. PLoS ONE. 2015;10(5):e0125600. 10.1371/journal.pone.0125600 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref036] 36.Tuno N, Tsuda Y, Takagi M. How zoophilic Japanese encephalitis vector mosquitoes feed on humans. J Med Entomol. 2017;54(1):8–13. 10.1093/jme/tjw165 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref037] 37.Eckhoff PA. A malaria transmission-directed model of mosquito life cycle and ecology. Malaria Journal. 2011;10:303. 10.1186/1475-2875-10-303 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref038] 38.Xu JC, Xu KQ, Li ZC, Meng FX, Tu TT, Xu L, et al. Forecast of dengue cases in 20 Chinese cities based on the deep learning method. Int J Environ Res Public Health. 2020;17(2):453. 10.3390/ijerph17020453 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0247980.ref039] 39.Jordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science. 2015;349(6245):255–60. 10.1126/science.aaa8415 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref040] 40.Freeman BS, Taylor G, Gharabaghi B, Thé J. Forecasting air quality time series using deep learning. J Air Waste Manag Assoc. 2018;68(8):866–86. 10.1080/10962247.2018.1459956 [DOI] [PubMed] [Google Scholar]

[pone.0247980.ref041] 41.Chae S, Kwon S, Lee D. Predicting infectious disease using deep learning and big data. Int J Environ Res Public Health. 2018;15(8):1596. 10.3390/ijerph15081596 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Association between meteorological factors and the prevalence dynamics of Japanese encephalitis

Taotian Tu

Keqiang Xu

Lei Xu

Yuan Gao

Ying Zhou

Yaming He

Yang Liu

Qiyong Liu

Hengqing Ji

Wenge Tang

Roles

Abstract

Introduction

Materials and methods

Study area

Fig 1. Spatial distribution of the study area and Japanese encephalitis (JE) cases.

Data collection

Table 1. The feature parameters used in this study.

Statistical modelling

Fig 2. Summarized workflow for the construction of the LSTM-based forecasting model for JE cases and its comparison with other candidate models.

Results

Outcomes of ZIGAM

Fig 3. Monthly time series plots in Chongqing from 2007 to 2019.

Fig 4. Analysis of potentially non-linear influences on the incidence of JE in Chongqing based on data from 2007–2019.

Outcomes of the LSTM model

Fig 5. Time series of the incidence of JE in Chongqing.

Discussion

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Ram K Raghavan

Roles

Author response to Decision Letter 0

Decision Letter 1

Ram K Raghavan

Roles

Author response to Decision Letter 1

Decision Letter 2

Ram K Raghavan

Roles

Acceptance letter

Ram K Raghavan

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases