Modeling pollen time series using seasonal-trend decomposition procedure based on LOESS smoothing

Unraveling the spatiotemporal dynamics and drivers of surface and tropospheric ozone in China

In China, the rapid development of the economy and the implementation of multiple mitigation strategies in recent decades have led to dramatic changes in air quality. In this study, ground- and satellite-based observations were integrated to comprehensively investigate the spatiotemporal variations of nationwide surface-level and tropospheric ozone (O3). Additionally, a meteorology correction model was developed, combining decomposition analysis and stepwise multiple linear regression, to explore the influence of meteorological conditions, anthropogenic emissions, and control actions on surface and tropospheric O3 in China. The results suggest that O3 pollution in China has strong spatial characteristics and seasonal patterns. The nationwide tropospheric O3 pollution substantially deteriorated from 2005 to 2020, and the increasing rate of tropospheric column O3 in the four target regions ranged from 0.20 (0.15-0.26) Dobson Units yr-1 to 0.26 (0.17-0.34) Dobson Units yr-1. Simultaneously, the meteorological factors exerted distinct influences on the surface-level and tropospheric O3 by regions. Since 2018, both surface and tropospheric O3 have declined remarkably in China. The meteorology correction model indicates that the downward trend is primarily attributed to the implementation of effective control plans (the Three-Year Action Plan for Cleaner Air) and the reduction of anthropogenic emissions. However, it is notable that surface-level O3 in autumn-winter, particularly in the eastern and southern part, increased markedly in recent years. These findings imply that the current mitigation strategy still has some insufficiencies, and to reduce the exposure risk in the future, China needs to set more ambitious mitigation targets and continue to push forward and strengthen the synergetic control of multiple air pollutants.

Exposure of insects to current use pesticide residues in soil and vegetation along spatial and temporal distribution in agricultural sites

Current use pesticides (CUPs) are recognised as the largest deliberate input of bioactive substances into terrestrial ecosystems and one of the main factors responsible for the current decline in insects in agricultural areas. To quantify seasonal insect exposure in the landscape at a regional scale (Rhineland-Palatine in Germany), we analysed the presence of multiple (93) active ingredients in CUPs across three different agricultural cultivation types (with each three fields: arable, vegetable, viticulture) and neighbouring meadows. We collected monthly soil and vegetation samples over a year. A total of 71 CUP residues in different mixtures was detected, with up to 28 CUPs in soil and 25 in vegetation in single samples. The concentrations and numbers of CUPs in vegetation fluctuated over the sampling period, peaking in the summer months in the vegetation but remaining almost constant in topsoil. We calculated in-field additive risks for earthworms, collembola, and soil-living wild bees using the measured soil concentrations of CUPs. Our results call for the need to assess CUP mixture risks at low concentrations, as multiple residues are chronically present in agricultural areas. Since this risk is not addressed in regulation, we emphasise the urgent need to implement global pesticide reduction targets.

Google trend analysis of the Indian population reveals a panel of seasonally sensitive comorbid symptoms with implications for monitoring the seasonally sensitive human population

Seasonal variations in the environment induce observable changes in the human physiological system and manifest as various clinical symptoms in a specific human population. Our earlier studies predicted four global severe seasonal sensitive comorbid lifestyle diseases (SCLDs), namely, asthma, obesity, hypertension, and fibrosis. Our studies further indicated that the SCLD category of the human population may be maladapted or unacclimatized to seasonal changes. The current study aimed to explore the major seasonal symptoms associated with SCLD and evaluate their seasonal linkages via Google Trends (GT). We used the Human Disease Symptom Network (HSDN) to dissect common symptoms of SCLD. We then exploited medical databases and medical literature resources in consultation with medical practitioners to narrow down the clinical symptoms associated with four SCLDs, namely, pulmonary hypertension, pulmonary fibrosis, asthma, and obesity. Our study revealed a strong association of 12 clinical symptoms with SCLD. Each clinical symptom was further subjected to GT analysis to address its seasonal linkage. The GT search was carried out in the Indian population for the period from January 2015-December 2019. In the GT analysis, 11 clinical symptoms were strongly associated with Indian seasonal changes, with the exception of hypergammaglobulinemia, due to the lack of GT data in the Indian population. These 11 symptoms also presented sudden increases or decreases in search volume during the two major Indian seasonal transition months, namely, March and November. Moreover, in addition to SCLD, several seasonally associated clinical disorders share most of these 12 symptoms. In this regard, we named these 12 symptoms the "seasonal sensitive comorbid symptoms (SSC)" of the human population. Further clinical studies are needed to verify the utility of these symptoms in screening seasonally maladapted human populations. We also warrant that clinicians and researcher be well aware of the limitations and pitfalls of GT before correlating the clinical outcome of SSC symptoms with GT.

Exploring the influence of meteorological conditions on the variability of olive pollen intradiurnal patterns: Differences between pre- and post-peak periods

Olive trees hold a significant economical, ecological and ornamental value, especially in the Mediterranean area. It is a wind-pollinated species emitting huge quantities of pollen with a high degree of allergenic sensitization. Andalusia region (southern Spain), where 15 % of the global olive tree population is cultivated, present a high density of this crop, reaching daily airborne olive pollen concentrations up to 6.000 pollen/m3. Although daily variations during the pollen season have been widely investigated in bibliography, factors influencing the intradiurnal dynamics of olive pollen concentrations remains underexplored in aerobiology. The present paper focuses on it, characterizing main intradiurnal patterns, identifying potential pollen source areas and the influence of wind dynamics on Córdoba city olive pollen data. The results reveal the presence of different pollen peaks at various hours of the day, depending on the stage of the pollen season (pre- and post-peak) and wind dynamics. Nevertheless, the main one is detected at midday during the pre-peak season, with secondary peaks at night, morning and late afternoon. A thorough examination of wind dynamics highlighted the significant influence of distant and local sources on the hourly pollen peaks and hence, on intradiurnal patterns. The analysis of the intradiurnal pattern associated with different air mass patterns demonstrated a considerable variability in the occurrence of peak concentrations and hence, in the contribution of sources. The characterization of surface winds confirms the substantial differences in the dynamics of atmospheric transport processes that influence the primary intradiurnal patterns of olive pollen in this region.

Zoonotic outbreak risk prediction with long short-term memory models: a case study with schistosomiasis, echinococcosis, and leptospirosis

BACKGROUND

Zoonotic infections, characterized with huge pathogen diversity, wide affecting area and great society harm, have become a major global public health problem. Early and accurate prediction of their outbreaks is crucial for disease control. The aim of this study was to develop zoonotic diseases risk predictive models based on time-series incidence data and three zoonotic diseases in mainland China were employed as cases.

METHODS

The incidence data for schistosomiasis, echinococcosis, and leptospirosis were downloaded from the Scientific Data Centre of the National Ministry of Health of China, and were processed by interpolation, dynamic curve reconstruction and time series decomposition. Data were decomposed into three distinct components: the trend component, the seasonal component, and the residual component. The trend component was used as input to construct the Long Short-Term Memory (LSTM) prediction model, while the seasonal component was used in the comparison of the periods and amplitudes. Finaly, the accuracy of the hybrid LSTM prediction model was comprehensive evaluated.

RESULTS

This study employed trend series of incidence numbers and incidence rates of three zoonotic diseases for modeling. The prediction results of the model showed that the predicted incidence number and incidence rate were very close to the real incidence data. Model evaluation revealed that the prediction error of the hybrid LSTM model was smaller than that of the single LSTM. Thus, these results demonstrate that using trending sequences as input sequences for the model leads to better-fitting predictive models.

CONCLUSIONS

Our study successfully developed LSTM hybrid models for disease outbreak risk prediction using three zoonotic diseases as case studies. We demonstrate that the LSTM, when combined with time series decomposition, delivers more accurate results compared to conventional LSTM models using the raw data series. Disease outbreak trends can be predicted more accurately using hybrid models.

The impact of grazing activities and environmental conditions on the stability of alpine grassland ecosystems

Globally, grazing activities have profound impacts on the structure and function of ecosystems. This study, based on a 20-year MODIS time series dataset, employs remote sensing techniques and the Seasonal-Trend decomposition using Loess (STL) algorithm to quantitatively assess the stability of alpine grassland ecosystems from multiple dimensions, and to reveal the characteristics of grazing activities and environmental conditions on ecosystem stability. The results indicate that only 5.77% of the area remains undisturbed, with most areas experiencing varying degrees of disturbance. Further analysis shows that grazing activities in high vegetation coverage areas are the main source of interference. In areas with concentrated interference, elevation and slope have a positive correlation with resistance stability, but a negative correlation with recovery stability. Precipitation and landscape diversity have positive effects on both resistance stability and recovery stability. Vegetation coverage and grazing intensity have a negative correlation with resistance stability, but a positive correlation with recovery stability. This highlights the complex interactions between human activities, environmental factors, and ecosystem stability. The findings emphasize the need for targeted conservation and management strategies to mitigate disturbances to ecosystems affected by human activities and enhance their stability.

The effect of power stretchers on occupational injury rates in an urban emergency medical services system

BACKGROUND

To examine occupational injury rates in a dual-response emergency medical services (EMS) system before and after implementation of a power-lift stretcher system.

METHODS

The seasonally-adjusted occupational injury rate was estimated relative to medical call volume (per 1000 calls) and workers (per 100 FTEs) from 2009 to 2019, and stratified by severity (lost-time, healthcare only), role (EMS, FIRE) and type (patient-handling). Power-lift stretchers were adopted between 2013 and 2015. Preinjury versus postinjury rates were compared using binomial tests. Interrupted time series (ITS) analysis was used to estimate the trend and change in injuries related to patient-handling, with occupational illnesses serving as control.

RESULTS

Binomial tests revealed varied results, with reductions in the injury rate per 1000 calls (-14.0%) and increases in the rate per 100 FTEs (+14.1%); rates also differed by EMS role and injury severity. ITS analysis demonstrated substantial reductions in patient-handling injuries following implementation of power-lift stretchers, both in the injury rate per 1000 calls (-50.4%) and per 100 FTEs (-46.6%), specifically among individuals deployed on the ambulance. Injury rates were slightly elevated during the winter months (+0.8 per 100 FTEs) and lower during spring (-0.5 per 100 FTEs).

CONCLUSIONS

These results support the implementation of power-lift stretchers for injury prevention in EMS systems and demonstrate advantages of ITS analysis when data span long preintervention and postintervention periods.

Association between ambient temperature and common allergenic pollen and fungal spores: A 52-year analysis in central England, United Kingdom

Exposure to pollen and fungal spores can trigger asthma/allergic symptoms and affect health. Rising temperatures from climate change have been associated with earlier seasons and increasing intensity for some pollen, with weaker evidence for fungal spores. It is unclear whether climate change has resulted in changes in the exposure-response function between temperature and pollen/fungal spore concentrations over time. This study examined associations between temperature and pollen/fungal spores in different time periods and assessed potential adaptation using the longest pollen/fungal spore dataset in existence (52 years). Daily concentrations of pollen (birch and grass) and fungal spores (Cladosporium, Alternaria, Sporobolomyces and Tilletiopsis) collected between April and October from Derby (1970-2005) and Leicester (2006-2021), UK, were analysed. Cumulative seasonal concentrations (seasonal integral) and start-of-season were calculated and linked to seasonal mean temperatures (Tmeans) using generalized additive models. Daily concentrations were evaluated against daily Tmean with distributed lagged nonlinear models. Models were adjusted for precipitation, relative humidity, long-term trend and location. Seasonal and daily analyses were respectively stratified into two periods (1970-1995, 1997-2021) and five decades. Warmer seasonal Tmeans were associated with higher seasonal integral for birch, Cladosporium and Alternaria, as well as earlier start-of-season for birch, grass and Cladosporium. There were indications of changing associations with temperature in the recent decades. A warmer January was associated with higher seasonal integral for grass in 1997-2021, but not in 1970-1995. In 2000-2021, daily concentrations of birch pollen tended to remain at higher levels, vs. decrease during 1990s, when Tmean was between 13 and 15 °C. Our study suggests higher temperatures experienced in recent decades are associated with higher overall abundance of some pollen/fungal spores, which may increase future disease burdens of allergies. The changing responses of some pollen to higher temperatures over time may indicate adaptation to increasing temperatures and should be considered in climate change mitigation and adaptation planning.

Correlation Recurrent Units: A Novel Neural Architecture for Improving the Predictive Performance of Time-Series Data

Time-series forecasting (TSF) is a traditional problem in the field of artificial intelligence, and models such as recurrent neural network, long short-term memory, and gate recurrent units have contributed to improving its predictive accuracy. Furthermore, model structures have been proposed to combine time-series decomposition methods such as seasonal-trend decomposition using LOESS. However, this approach is learned in an independent model for each component, and therefore, it cannot learn the relationships between the time-series components. In this study, we propose a new neural architecture called a correlation recurrent unit (CRU) that can perform time-series decomposition within a neural cell and learn correlations (autocorrelation and correlation) between each decomposition component. The proposed neural architecture was evaluated through comparative experiments with previous studies using four univariate and four multivariate time-series datasets. The results showed that long- and short-term predictive performance was improved by more than 10%. The experimental results indicate that the proposed CRU is an excellent method for TSF problems compared to other neural architectures.

Psychiatric medication prescriptions increasing for college students above and beyond the COVID-19 pandemic

Psychiatric medication prescriptions for college students have been rising since 2007, with approximately 17% of college students prescribed medication for a mental health issue. This increase mirrors overall increases in both mental health diagnoses and treatment of university students. As psychiatric medication prescriptions for college students were increasing prior to pandemic, the goal of this study was to compare these prescriptions over the years, while accounting for the added stressor of the COVID-19 pandemic. This study utilized cross-sectional, retrospective data from a cohort of college students receiving care from the university's health service. We examined prescriptions for mental healthcare from 2015 to 2021. There was a significant increase in the percentage of psychiatric medication prescriptions in 2020 (baseline 15.8%; threshold 3.5%) and 2021 (baseline 41.3%; threshold 26.3%) compared to the historical baseline average for the whole sample and as well as for female students (2020 baseline 21.3% and threshold 4.6%; 2021 baseline 55.1% and threshold 33.7%). Within these years, we found higher trends for prescriptions in April-May as well as September-December. Overall, we found that psychiatric medication prescriptions have continued to rise through the years, with a large increase occurring during the pandemic. In addition, we found that these increases reflect the academic year, which is important for university health centers to consider when they are planning to staff clinics and plan the best way to treat college students with mental health difficulties in the future.

A study of the long term changes in the electromagnetic environment using data from continuous monitoring sensors in Greece

Owing to the advancement of wireless technologies, there is a strong public perception of increasing exposure to Radiofrequency (RF) electromagnetic fields (EMF). The aim of this study is to determine the evolution of EMF in the environment, and consequently, human exposure to them, over a period of about two decades, spanning from the end of 2003 until February 2022. The study is based on data collected by two non-ionizing radiation monitoring networks in Greece. The networks consist of fixed EMF sensors that register the RMS electric field value every 6 min, on a 24 h basis. We used the Seasonal-Trend decomposition method using (LOESS), known as the STL method to decompose the time series into trend, seasonal, and noise components. Additionally, since the sensors include frequency filters for separating the cellular frequencies, the recorded data were used to identify the exposure contribution by cellular networks in comparison to other EMF sources. The study indicates that RF-EMF do not explicitly decrease or increase but rather fluctuate over time. Similarly, the contribution of mobile cellular networks to the total field change over time.

Isolating the species element in grass pollen allergy: A review

Grass pollen is a leading cause of allergy in many countries, particularly Europe. Although many elements of grass pollen production and dispersal are quite well researched, gaps still remain around the grass species that are predominant in the air and which of those are most likely to trigger allergy. In this comprehensive review we isolate the species aspect in grass pollen allergy by exploring the interdisciplinary interdependencies between plant ecology, public health, aerobiology, reproductive phenology and molecular ecology. We further identify current research gaps and provide open ended questions and recommendations for future research in an effort to focus the research community to develop novel strategies to combat grass pollen allergy. We emphasise the role of separating temperate and subtropical grasses, identified through divergence in evolutionary history, climate adaptations and flowering times. However, allergen cross-reactivity and the degree of IgE connectivity in sufferers between the two groups remains an area of active research. The importance of future research to identify allergen homology through biomolecular similarity and the connection to species taxonomy and practical implications of this to allergenicity is further emphasised. We also discuss the relevance of eDNA and molecular ecological techniques (DNA metabarcoding, qPCR and ELISA) as important tools in quantifying the connection between the biosphere with the atmosphere. By gaining more understanding of the connection between species-specific atmospheric eDNA and flowering phenology we will further elucidate the importance of species in releasing grass pollen and allergens to the atmosphere and their individual role in grass pollen allergy.

Hybrid approaches for container traffic forecasting in the context of anomalous events: The case of the Yangtze River Delta region in the COVID-19 pandemic

The COVID-19 pandemic had a significant impact on container transportation. Accurate forecasting of container throughput is critical for policymakers and port authorities, especially in the context of the anomalous events of the COVID-19 pandemic. In this paper, we firstly proposed hybrid models for univariate time series forecasting to enhance prediction accuracy while eliminating the nonlinearity and multivariate limitations. Next, we compared the forecasting accuracy of different models with various training dataset extensions and forecasting horizons. Finally, we analysed the impact of the COVID-19 pandemic on container throughput forecasting and container transportation. An empirical analysis of container throughputs in the Yangtze River Delta region was performed for illustration and verification purposes. Error metrics analysis suggests that SARIMA-LSTM₂ and SARIMA-SVR₂ (configuration 2) have the best performance compared to other models and they can better predict the container traffic in the context of anomalous events such as the COVID-19 pandemic. The results also reveal that, with an increase in the training dataset extensions, the accuracy of the models is improved, particularly in comparison with standard statistical models (i.e. SARIMA model). An accurate prediction can help strategic management and policymakers to better respond to the negative impact of the COVID-19 pandemic.

Data mining assessment of Poaceae pollen influencing factors and its environmental implications

Poaceae pollen is highly allergenic, with a marked contribution to the pollen worldwide allergy prevalence. Pollen counts are defined by the species present in the considered area, although year-to-year oscillations may be triggered by different parameters, among which are weather conditions. Due to the predominant role of Poaceae pollen in the allergenicity in urban green areas, the aim of this study was the analysis of pollen trends and the influence of meteorology to forecast relevant variations in airborne pollen levels. The study was carried out during the 1993-2020 period in Ourense, in NW Iberian Peninsula. We used a volumetric Lanzoni VPPS 2000 trap for recording Poaceae airborne pollen grains, and meteorological daily data were obtained from the Galician Institute for Meteorology and Oceanography. The main indexes of the pollen season and their trends were calculated. A correlation analysis and 'C5.0 Decision Trees and Rule-Based Models' data mining algorithm were applied to determine the influence of meteorological conditions on pollen levels. We detected atmospheric Poaceae pollen during 139 days on average, mainly from April to August. The mean pollen grains amount recorded during the pollen season was 4608 pollen grains, with the pollen maximum peak of 276 pollen/m³ on 27 June. We found no statistically significant trends and slight slopes for the seasonal indexes, similarly to previous Poaceae studies in the same region. The calculated C5.0 model offered defined results, indicating that the combination of mean temperature above 17.46 °C and sunlight exposure higher than 12.7 h is conductive to significantly high pollen levels. The obtained results make possible the identification of risk moments during the pollen season for the activation of protective measures for sensitized population to grass pollen.

Atmospheric transport reveals grass pollen dispersion distances

Identifying the origin of bioaerosols is of central importance in many biological disciplines, such as human health, agriculture, forestry, aerobiology and conservation. Modelling sources, transportation pathways and sinks can reveal how bioaerosols vary in the atmosphere and their environmental impact. Grass pollen are particularly important due to their widely distributed source areas, relatively high abundance in the atmosphere and high allergenicity. Currently, studies are uncertain regarding sampler representability between distance and sources for grass pollen. Using generalized linear modelling, this study aimed to analyse this relationship further by answering the question of distance-to-source area contribution. Grass pollen concentrations were compared between urban and rural locations, located 6.4 km apart, during two years in Worcestershire, UK. We isolated and refined vegetation areas at 100 m × 100 m using the 2017 CEH Crop Map and conducted atmospheric modelling using HYSPLIT to identify which source areas could contribute pollen. Pollen concentrations were then modelled with source areas and meteorology using generalized linear mixed-models with three temporal variables as random variation. We found that the Seasonal Pollen Integral for grass pollen varied between both years and location, with the urban location having higher levels. Day of year showed higher temporal variation than the diurnal or annual variables. For the urban location, grass source areas within 30 km had positive significant effects in predicting grass pollen concentrations, while source areas within 2-10 km were important for the rural one. The source area differential was likely influenced by an urban-rural gradient that caused differences in the source area contribution. Temperature had positive highly significant effects on both locations while precipitation affected only the rural location. Combining atmospheric modelling, vegetation source maps and generalized linear modelling was found to be a highly accurate tool to identify transportation pathways of bioaerosols in landscape environments.

Global evidence of expressed sentiment alterations during the COVID-19 pandemic

The COVID-19 pandemic has created unprecedented burdens on people's physical health and subjective well-being. While countries worldwide have developed platforms to track the evolution of COVID-19 infections and deaths, frequent global measurements of affective states to gauge the emotional impacts of pandemic and related policy interventions remain scarce. Using 654 million geotagged social media posts in over 100 countries, covering 74% of world population, coupled with state-of-the-art natural language processing techniques, we develop a global dataset of expressed sentiment indices to track national- and subnational-level affective states on a daily basis. We present two motivating applications using data from the first wave of COVID-19 (from 1 January to 31 May 2020). First, using regression discontinuity design, we provide consistent evidence that COVID-19 outbreaks caused steep declines in expressed sentiment globally, followed by asymmetric, slower recoveries. Second, applying synthetic control methods, we find moderate to no effects of lockdown policies on expressed sentiment, with large heterogeneity across countries. This study shows how social media data, when coupled with machine learning techniques, can provide real-time measurements of affective states.

Short-Term River Flood Forecasting Using Composite Models and Automated Machine Learning: The Case Study of Lena River

The paper presents a hybrid approach for short-term river flood forecasting. It is based on multi-modal data fusion from different sources (weather stations, water height sensors, remote sensing data). To improve the forecasting efficiency, the machine learning methods and the Snowmelt-Runoff physical model are combined in a composite modeling pipeline using automated machine learning techniques. The novelty of the study is based on the application of automated machine learning to identify the individual blocks of a composite pipeline without involving an expert. It makes it possible to adapt the approach to various river basins and different types of floods. Lena River basin was used as a case study since its modeling during spring high water is complicated by the high probability of ice-jam flooding events. Experimental comparison with the existing methods confirms that the proposed approach reduces the error at each analyzed level gauging station. The value of Nash–Sutcliffe model efficiency coefficient for the ten stations chosen for comparison is 0.80. The other approaches based on statistical and physical models could not surpass the threshold of 0.74. Validation for a high-water period also confirms that a composite pipeline designed using automated machine learning is much more efficient than stand-alone models.

Modeling of Diurnal Changing Patterns in Airborne Crop Remote Sensing Images

Airborne remote sensing technologies have been widely applied in field crop phenotyping. However, the quality of current remote sensing data suffers from significant diurnal variances. The severity of the diurnal issue has been reported in various plant phenotyping studies over the last four decades, but there are limited studies on the modeling of the diurnal changing patterns that allow people to precisely predict the level of diurnal impacts. In order to comprehensively investigate the diurnal variability, it is necessary to collect time series field images with very high sampling frequencies, which has been difficult. In 2019, Purdue agricultural (Ag) engineers deployed their first field visible to near infrared (VNIR) hyperspectral gantry platform, which is capable of repetitively imaging the same field plots every 2.5 min. A total of 8631 hyperspectral images of the same field were collected for two genotypes of corn plants from the vegetative stage V4 to the reproductive stage R1 in the 2019 growing season. The analysis of these images showed that although the diurnal variability is very significant for almost all the image-derived phenotyping features, the diurnal changes follow stable patterns. This makes it possible to predict the imaging drifts by modeling the changing patterns. This paper reports detailed diurnal changing patterns for several selected plant phenotyping features such as Normalized Difference Vegetation Index (NDVI), Relative Water Content (RWC), and single spectrum bands. For example, NDVI showed a repeatable V-shaped diurnal pattern, which linearly drops by 0.012 per hour before the highest sun angle and increases thereafter by 0.010 per hour. The different diurnal changing patterns in different nitrogen stress treatments, genotypes and leaf stages were also compared and discussed. With the modeling results of this work, Ag remote sensing users will be able to more precisely estimate the deviation/change of crop feature predictions caused by the specific imaging time of the day. This will help people to more confidently decide on the acceptable imaging time window during a day. It can also be used to calibrate/compensate the remote sensing result against the time effect.

Predicting the Olea pollen concentration with a machine learning algorithm ensemble

Air pollution in large cities produces numerous diseases and even millions of deaths annually according to the World Health Organization. Pollen exposure is related to allergic diseases, which makes its prediction a valuable tool to assess the risk level to aeroallergens. However, airborne pollen concentrations are difficult to predict due to the inherent complexity of the relationships among both biotic and environmental variables. In this work, a stochastic approach based on supervised machine learning algorithms was performed to forecast the daily Olea pollen concentrations in the Community of Madrid, central Spain, from 1993 to 2018. Firstly, individual Light Gradient Boosting Machine (LightGBM) and artificial neural network (ANN) models were applied to predict the day of the year (DOY) when the peak of the pollen season occurs, resulting the estimated average peak date 149.1 ± 9.3 and 150.1 ± 10.8 DOY for LightGBM and ANN, respectively, close to the observed value (148.8 ± 9.8). Secondly, the daily pollen concentrations during the entire pollen season have been calculated using an ensemble of two-step GAM followed by LightGBM and ANN. The results of the prediction of daily pollen concentrations showed a coefficient of determination (r²) above 0.75 (goodness of the model following cross-validation). The predictors included in the ensemble models were meteorological variables, phenological metrics, specific site-characteristics, and preceding pollen concentrations. The models are state-of-the-art in machine learning and their potential has been shown to be used and deployed to understand and to predict the pollen risk levels during the main olive pollen season.

Deep Learning Framework with Time Series Analysis Methods for Runoff Prediction

Recent advances in deep learning, especially the long short-term memory (LSTM) networks, provide some useful insights on how to tackle time series prediction problems, not to mention the development of a time series model itself for prediction. Runoff forecasting is a time series prediction problem with a series of past runoff data (water level and discharge series data) as inputs and a fixed-length series of future runoff as output. Most previous work paid attention to the sufficiency of input data and the structural complexity of deep learning, while less effort has been put into the consideration of data quantity or the processing of original input data—such as time series decomposition, which can better capture the trend of runoff—or unleashing the effective potential of deep learning. Mutual information and seasonal trend decomposition are two useful time series methods in handling data quantity analysis and original data processing. Based on a former study, we proposed a deep learning model combined with time series analysis methods for daily runoff prediction in the middle Yangtze River and analyzed its feasibility and usability with frequently used counterpart models. Furthermore, this research also explored the data quality that affect the performance of the deep learning model. With the application of the time series method, we can effectively get some information about the data quality and data amount that we adopted in the deep learning model. The comparison experiment resulted in two different sites, implying that the proposed model improved the precision of runoff prediction and is much easier and more effective for practical application. In short, time series analysis methods can exert great potential of deep learning in daily runoff prediction and may unleash great potential of artificial intelligence in hydrology research.

Real-Time Traffic Flow Forecasting via a Novel Method Combining Periodic-Trend Decomposition

Accurate and timely traffic flow forecasting is a critical task of the intelligent transportation system (ITS). The predicted results offer the necessary information to support the decisions of administrators and travelers. To investigate trend and periodic characteristics of traffic flow and develop a more accurate prediction, a novel method combining periodic-trend decomposition (PTD) is proposed in this paper. This hybrid method is based on the principle of “decomposition first and forecasting last”. The well-designed PTD approach can decompose the original traffic flow into three components, including trend, periodicity, and remainder. The periodicity is a strict period function and predicted by cycling, while the trend and remainder are predicted by modelling. To demonstrate the universal applicability of the hybrid method, four prevalent models are separately combined with PTD to establish hybrid models. Traffic volume data are collected from the Minnesota Department of Transportation (Mn/DOT) and used to conduct experiments. Empirical results show that the mean absolute error (MAE), mean absolute percentage error (MAPE), and mean square error (MSE) of hybrid models are averagely reduced by 17%, 17%, and 29% more than individual models, respectively. In addition, the hybrid method is robust for a multi-step prediction. These findings indicate that the proposed method combining PTD is promising for traffic flow forecasting.

Deep Learning Predictor for Sustainable Precision Agriculture Based on Internet of Things System

Based on the collected weather data from the agricultural Internet of Things (IoT) system, changes in the weather can be obtained in advance, which is an effective way to plan and control sustainable agricultural production. However, it is not easy to accurately predict the future trend because the data always contain complex nonlinear relationship with multiple components. To increase the prediction performance of the weather data in the precision agriculture IoT system, this study used a deep learning predictor with sequential two-level decomposition structure, in which the weather data were decomposed into four components serially, then the gated recurrent unit (GRU) networks were trained as the sub-predictors for each component. Finally, the results from GRUs were combined to obtain the medium- and long-term prediction result. The experiments were verified for the proposed model based on weather data from the IoT system in Ningxia, China, for wolfberry planting, in which the prediction results showed that the proposed predictor can obtain the accurate prediction of temperature and humidity and meet the needs of precision agricultural production.

Deep Hybrid Model Based on EMD with Classification by Frequency Characteristics for Long-Term Air Quality Prediction

Air pollution (mainly PM2.5) is one of the main environmental problems about air quality. Air pollution prediction and early warning is a prerequisite for air pollution prevention and control. However, it is not easy to accurately predict the long-term trend because the collected PM2.5 data have complex nonlinearity with multiple components of different frequency characteristics. This study proposes a hybrid deep learning predictor, in which the PM2.5 data are decomposed into components by empirical mode decomposition (EMD) firstly, and a convolutional neural network (CNN) is built to classify all the components into a fixed number of groups based on the frequency characteristics. Then, a gated-recurrent-unit (GRU) network is trained for each group as the sub-predictor, and the results from the three GRUs are fused to obtain the prediction result. Experiments based on the PM2.5 data from Beijing verify the proposed model, and the prediction results show that the decomposition and classification can develop the accuracy of the proposed predictor for air pollution prediction greatly.

Integrated Predictor Based on Decomposition Mechanism for PM2.5 Long-Term Prediction

It is crucial to predict PM2.5 concentration for early warning regarding and the control of air pollution. However, accurate PM2.5 prediction has been challenging, especially in long-term prediction. PM2.5 monitoring data comprise a complex time series that contains multiple components with different characteristics; therefore, it is difficult to obtain an accurate prediction by a single model. In this study, an integrated predictor is proposed, in which the original data are decomposed into three components, that is, trend, period, and residual components, and then different sub-predictors including autoregressive integrated moving average (ARIMA) and two gated recurrent units are used to separately predict the different components. Finally, all the predictions from the sub-predictors are combined in fusion node to obtain the final prediction for the original data. The results of predicting the PM2.5 time series for Beijing, China showed that the proposed predictor can effectively improve prediction accuracy for long-term prediction.

A descriptive analysis of the Spatio-temporal distribution of intestinal infectious diseases in China

BACKGROUND

Intestinal infectious diseases (IIDs) have caused numerous deaths worldwide, particularly among children. In China, eight IIDs are listed as notifiable infectious diseases, including cholera, poliomyelitis, dysentery, typhoid and paratyphoid (TAP), viral Hepatitis A, viral Hepatitis E, hand-foot-mouth disease (HFMD) and other infectious diarrhoeal diseases (OIDDs). The aim of the study is to analyse the spatio-temporal distribution of IIDs from 2006 to 2016.

METHODS

Data on the incidence of IIDs from 2006 to 2016 were collected from the public health science data centre issued by the Chinese Center for Disease Control and Prevention. This study applied seasonal decomposition analysis, spatial autocorrelation analysis and space-time scan analysis. Plots and maps were constructed to visualize the spatio-temporal distribution of IIDs.

RESULTS

Regarding temporal analysis, the incidence of HFMD and Hepatitis E showed a distinct increasing trend, while the incidence of TAP, dysentery, and Hepatitis A presented decreasing trends over the last decade. The incidence of OIID remained steady. Summer is the season with the greatest number of cases of different IIDs. Regarding the spatial distribution, approximately all p values for the global Moran's I from 2006 to 2016 were less than 0.05, indicating that the incidences of the epidemics were unevenly distributed throughout the country. The high-risk areas for HFMD and OIDD were located in the Beijing-Tianjin-Tangshan (BTT) region and south China. The high-risk areas for TAP were located in some parts of southwest China. A higher incidence rates for dysentery and Hepatitis A were observed in the BTT region and some west provincial units. The high-risk areas for Hepatitis E were the BTT region and the Yangtze River Delta area.

CONCLUSIONS

Based on our temporal and spatial analysis of IIDs, we identified the high-risk periods and clusters of regions for the diseases. HFMD and OIDD exhibited high incidence rates, which reflected the negligence of Class C diseases by the government. At the same time, the incidence rate of Hepatitis E gradually surpassed Hepatitis A. The authorities should pay more attention to Class C diseases and Hepatitis E. Regardless of the various distribution patterns of IIDs, disease-specific, location-specific, and disease-combined interventions should be established.

Health Assessment and Fault Detection System for an Industrial Robot Using the Rotary Encoder Signal

In an industrial robot, rotary encoders have been extensively used for dynamic control and positioning. This study shows that the encoder signal, after appropriate processing, can also be efficiently utilized for the health observation of energy performance of industrial robots system. Singular spectrum analysis (SSA) and Hilbert transform (HT) is proposed in this work, for detecting weak position oscillations to estimate the instantaneous amplitudes (IA) and the instantaneous frequencies (IF) of an industrial robot based on the encoder signal. Compared with empirical mode decomposition (EMD) and HT, the singular spectrum analysis and Hilbert transform (SSAHT) outperforms empirical mode decomposition Hilbert transform (EMDHT) in terms of ability and precision to determine source noise, and it can accurately catch the weak oscillations without signal deformation in both position and speed introduced via mechanical flaws. Combined with SSA, the IA and IF of both oscillations and residual are extracted by HT. They are obtained from the robot arm movement. These features play an important role in improving the performance detecting weak oscillations and the residual, essential information to evaluate the health conditions and fault detection to serve the energy performance for the industrial robot. The efficiency of the proposed system has been verified both numerical simulation and experimental data. The outcomes prove that the proposed SSAHT can detect flaw indications and additionally, it can also identify faulty components. Thus, the study presents a promising tool for the health monitoring of an industrial robot instead of the vibration-based monitoring scheme.

Standardised index for measuring atmospheric grass-pollen emission

Grass pollen is the main cause of pollen allergy in Europe, and-given its marked allergenic potential and elevated airborne concentrations-constitutes a major public health risk. This study sought to identify the grass species triggering allergies during the highest-risk periods, and to measure the contribution of each species to airborne grass pollen concentrations. This type of research is particularly useful with a view to optimising the prevention and diagnosis of pollen allergies and developing the most effective immunological treatments. To that end, a total of 28 species potentially responsible for allergies were analysed. In order to assess the potential contribution of these species to overall airborne pollen concentrations, an index was designed (Pollen Contribution Index) based on the following parameters for each species: flowering phenology, pollen grain size (polar and equatorial axes), abundance of the species in the area and pollen production. The species contributing most to airborne pollen concentrations were, in order: Dactylis glomerata subsp. hispanica, Lolium rigidum, Trisetum paniceum and Arrhenatherum album. These species all shared certain features: small grain size (and thus greater buoyancy in air), high pollen production and considerable abundance. This Index was applied to a case study in a Mediterranean-climate area of the central Iberian Peninsula, but could equally be applied to other areas and other allergenic pollens. Findings showed that a small number of species were responsible for most airborne grass pollen.

Epidemiological analysis of hemorrhagic fever with renal syndrome in China with the seasonal-trend decomposition method and the exponential smoothing model

Hemorrhagic fever with renal syndrome (HFRS) is one of the most common infectious diseases globally. With the most reported cases in the world, the epidemic characteristics are still remained unclear in China. This paper utilized the seasonal-trend decomposition (STL) method to analyze the periodicity and seasonality of the HFRS data, and used the exponential smoothing model (ETS) model to predict incidence cases from July to December 2016 by using the data from January 2006 to June 2016. Analytic results demonstrated a favorable trend of HFRS in China, and with obvious periodicity and seasonality, the peak of the annual reported cases in winter concentrated on November to January of the following year, and reported in May and June also constituted another peak in summer. Eventually, the ETS (M, N and A) model was adopted for fitting and forecasting, and the fitting results indicated high accuracy (Mean absolute percentage error (MAPE) = 13.12%). The forecasting results also demonstrated a gradual decreasing trend from July to December 2016, suggesting that control measures for hemorrhagic fever were effective in China. The STL model could be well performed in the seasonal analysis of HFRS in China, and ETS could be effectively used in the time series analysis of HFRS in China.