Dynamically pre-trained deep recurrent neural networks using environmental monitoring data for predicting PM2.5

Air pollutant prediction model based on transfer learning two-stage attention mechanism

Atmospheric pollution significantly impacts the regional economy and human health, and its prediction has been increasingly emphasized. The performance of traditional prediction methods is limited due to the lack of historical data support in new atmospheric monitoring sites. Therefore, this paper proposes a two-stage attention mechanism model based on transfer learning (TL-AdaBiGRU). First, the first stage of the model utilizes a temporal distribution characterization algorithm to segment the air pollutant sequences into periods. It introduces a temporal attention mechanism to assign self-learning weights to the period segments in order to filter out essential period features. Then, in the second stage of the model, a multi-head external attention mechanism is introduced to mine the network's hidden layer key features. Finally, the adequate knowledge learned by the model at the source domain site is migrated to the new site to improve the prediction capability of the new site. The results show that (1) the model is modeled from the data distribution perspective, and the critical information within the sequence of periodic segments is mined in depth. (2) The model employs a unique two-stage attention mechanism to capture complex nonlinear relationships in air pollutant data. (3) Compared with the existing models, the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of the model decreased by 14%, 13%, and 4%, respectively, and the prediction accuracy was greatly improved.

A physics-informed long short-term memory (LSTM) model for estimating ammonia emissions from dairy manure during storage

Manure management on dairy farms impacts how farmers maximize its value as fertilizer, reduce operating costs, and minimize environmental pollution potential. A persistent challenge on many farms is minimizing ammonia losses through volatilization during storage to maintain manure nitrogen content. Knowing the quantities of emitted pollutants is at the core of designing and improving mitigation strategies for livestock operations. Although process-based models have improved the accuracy of estimating ammonia emissions, complex systems such as manure storage still need to be solved because some underlying science still needs work. This study presents a novel physics-informed long short-term memory (PI-LSTM) modeling approach combining traditional process-based with recurrent neural networks to estimate ammonia loss from dairy manure during storage. The method entails inverse modeling to optimize hyperparameters to improve the accuracy of estimating physicochemical properties pertinent to ammonia's transport and surface emissions. The study used open data sets from two on-farm studies on liquid dairy manure storage in Switzerland and Indiana, U.S.A. The root mean square errors were 1.51 g m-2 h-1 for the PI-LSTM model, 3.01 g m-2 h-1 for the base compartmental process-based (Base-CPBM) model, and 2.17 g m-2 h-1 for the hyperparameter-tuned compartmental process-based (HT-CPBM) model. In addition, the PI-LSTM model outperformed the Base-CPBM and the HT-CPBM models by 20 to 80 % during summer and spring, when most annual ammonia emissions occur. The study demonstrated that incorporating physical knowledge into machine learning models improves generalization accuracy. The outcomes of this study provide the scientific basis to improve policymaking decisions and the design of suitable on-farm strategies to minimize manure nutrient losses on dairy farms during storage periods.

Modeling the air pollution process using a novel multi-site and multi-scale method with adaptive utilization of spatio-temporal information

This study focuses on modeling air quality with an adaptive utilization of spatio-temporal information from multiple air quality monitoring stations under a multi-scale framework. To this end, it is necessary to consider different strategies to combine different methods to decompose the given series and to fuse multi-site information. Based on a systematic comparative analysis, we propose a novel multi-scale and multi-site modeling method named the multivariate empirical mode decomposition and spatial cosine-attention-based long short-term memory (MEMD-SCA-LSTM). The MEMD-SCA-LSTM first employs MEMD to decompose the multi-site air quality series into the scale-aligned components and then models the components at different scales. The high-frequency components are modeled by a novel SCA-LSTM, which employs LSTM with residual blocks to extract the temporal information and utilizes an attention mechanism based on the cosine similarity to adaptively extract interactions among different sites. Other relatively regular components are modeled by the LSTM. Empirical study on PM2.5 in Hong Kong has demonstrated the effectiveness of fusing multi-site information using the spatial attention (SA) mechanism under the multi-scale framework with MEMD. The proposed MEMD-SCA-LSTM can improve the one-day ahead modeling performance with the mean absolute error and the root mean square error reduced over 10%, compared to the baseline modeling methods. For the two-day and three-day ahead performance, the MEMD-SCA-LSTM is still the best one. Furthermore, by visualizing the attention weights, we illustrate that our proposed SCA-LSTM can overcome some limitations of the traditional attention mechanisms and that the attention weights exhibit more informative patterns which could be used to analysis the transport of air pollutant between sites. The proposed modeling method is a general method, which is feasible and applicable to other pollutants in other cities or regions.

Machine learning approaches reveal highly heterogeneous air quality co-benefits of the energy transition

Estimating health benefits of reducing fossil fuel use from improved air quality provides important rationales for carbon emissions abatement. Simulating pollution concentration is a crucial step of the estimation, but traditional approaches often rely on complicated chemical transport models that require extensive expertise and computational resources. In this study, we develop a machine learning framework that is able to provide precise and robust annual average fine particle (PM2.5) concentration estimations directly from a high-resolution fossil energy use dataset. Applications of the framework with Chinese data reveal highly heterogeneous health benefits of avoiding premature mortality by reducing fossil fuel use in different sectors and regions in China with a mean of $19/tCO2 and a standard deviation of $38/tCO2. Reducing rural and residential coal use offers the highest co-benefits with a mean of $151/tCO2. Our findings prompt careful policy designs to maximize cost-effectiveness in the transition toward a carbon-neutral energy system.

Machine learning algorithms to forecast air quality: a survey

Air pollution is a risk factor for many diseases that can lead to death. Therefore, it is important to develop forecasting mechanisms that can be used by the authorities, so that they can anticipate measures when high concentrations of certain pollutants are expected in the near future. Machine Learning models, in particular, Deep Learning models, have been widely used to forecast air quality. In this paper we present a comprehensive review of the main contributions in the field during the period 2011-2021. We have searched the main scientific publications databases and, after a careful selection, we have considered a total of 155 papers. The papers are classified in terms of geographical distribution, predicted values, predictor variables, evaluation metrics and Machine Learning model.

Spatiotemporal air quality forecasting and health risk assessment over smart city of NEOM

Modeling and predicting air pollution concentrations is important to provide early warnings about harmful atmospheric substances. However, uncertainty in the dynamic process and limited information about chemical constituents and emissions sources make air-quality predictions very difficult. This study proposed a novel deep-learning method to extract high levels of abstraction in data and capture spatiotemporal features at hourly and daily time intervals in NEOM City, Saudi Arabia. The proposed method integrated a residual network (ResNet) with the convolutional long short-term memory (ConvLSTM). The ConvLSTM method was boosted by a ResNet model for deeply extracting the spatial features from meteorological and pollutant data and thereby mitigating the loss of feature information. Then, health risk assessment was put forward to evaluate PM₁₀ and PM_2.5 risk sensitivity in five districts in NEOM City. Results revealed that the proposed method with effective feature extraction could greatly optimize the accuracy of spatiotemporal air quality forecasts compared to existing state-of-the-art models. For the next hour prediction tasks, the PM₁₀ and PM_2.5 of MASE were 9.13 and 13.57, respectively. The proposed method provides an effective solution to improve the prediction of air-pollution concentrations while being portable to other regions around the world.

Air quality measurement, prediction and warning using transfer learning based IOT system for ambient assisted living

Purpose

Indoor air quality monitoring is extremely important in urban, industrial areas. Considering the devastating effect of declining quality of air in major part of the countries like India and China, it is highly recommended to monitor the quality of air which can help people with respiratory diseases, children and elderly people to take necessary precautions and stay safe at their homes. The purpose of this study is to detect air quality and perform predictions which could be part of smart home automation with the use of newer technology.

Design/methodology/approach

This study proposes an Internet-of-Things (IoT)-based air quality measurement, warning and prediction system for ambient assisted living. The proposed ambient assisted living system consists of low-cost air quality sensors and ESP32 controller with new generation embedded system architecture. It can detect Indoor Air Quality parameters like CO, PM2.5, NO2, O3, NH3, temperature, pressure, humidity, etc. The low cost sensor data are calibrated using machine learning techniques for performance improvement. The system has a novel prediction model, multiheaded convolutional neural networks-gated recurrent unit which can detect next hour pollution concentration. The model uses a transfer learning (TL) approach for prediction when the system is new and less data available for prediction. Any neighboring site data can be used to transfer knowledge for early predictions for the new system. It can have a mobile-based application which can send warning notifications to users if the Indoor Air Quality parameters exceed the specified threshold values. This is all required to take necessary measures against bad air quality.

Findings

The IoT-based system has implemented the TL framework, and the results of this study showed that the system works efficiently with performance improvement of 55.42% in RMSE scores for prediction at new target system with insufficient data.

Originality/value

This study demonstrates the implementation of an IoT system which uses low-cost sensors and deep learning model for predicting pollution concentration. The system is tackling the issues of the low-cost sensors for better performance. The novel approach of pretrained models and TL work very well at the new system having data insufficiency issues. This study contributes significantly with the usage of low-cost sensors, open-source advanced technology and performance improvement in prediction ability at new systems. Experimental results and findings are disclosed in this study. This will help install multiple new cost-effective monitoring stations in smart city for pollution forecasting.

Comparison of machine learning and deep learning techniques for the prediction of air pollution: a case study from China

The adverse effect of air pollution has always been a problem for human health. The presence of a high level of air pollutants can cause severe illnesses such as emphysema, chronic obstructive pulmonary disease (COPD), or asthma. Air quality prediction helps us to undertake practical action plans for controlling air pollution. The Air Quality Index (AQI) reflects the degree of concentration of pollutants in a locality. The average AQI was calculated for the various cities in China to understand the annual trends. Furthermore, the air quality index has been predicted for ten major cities across China using five different deep learning techniques, namely, Recurrent Neural Network (RNN), Bidirectional Gated Recurrent unit (Bi-GRU), Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Network BiLSTM (CNN-BiLSTM), and Convolutional BiLSTM (Conv1D-BiLSTM). The performance of these models has been compared with a machine learning model, eXtreme Gradient Boosting (XGBoost) to discover the most efficient deep learning model. The results suggest that the machine learning model, XGBoost, outperforms the deep learning models. While Conv1D-BiLSTM and CNN-BiLSTM perform well among the deep learning models in the estimation of the air quality index (AQI), RNN and Bi-GRU are the least performing ones. Thus, both XGBoost and neural network models are capable of capturing the non-linearity present in the dataset with reliable accuracy.

Prediction and assessment of the impact of COVID-19 lockdown on air quality over Kolkata: a deep transfer learning approach

The present study focuses on the prediction and assessment of the impact of lockdown because of coronavirus pandemic on the air quality during three different phases, viz., normal periods (1 January 2018-23 March 2020), complete lockdown (24 March 2020-31 May 2020), and partial lockdown (1 June 2020-30 September 2020). We identify the most important air pollutants influencing the air quality of Kolkata during three different periods using Random Forest, a tree-based machine learning (ML) algorithm. It is found that the ambient air quality of Kolkata is mainly affected with the aid of particulate matter or PM (PM₁₀ and PM_2.5). However, the effect of the lockdown is most prominent on PM_2.5 which spreads in the air of Kolkata due to diesel-driven vehicles, domestic and commercial combustion activities, road dust, and open burning. To predict urban PM_2.5 and PM₁₀ concentrations 24 h in advance, we use a deep learning (DL) model, namely, stacked-bidirectional long short-term memory (stacked-BDLSTM). The model is trained during the normal periods, and it shows the superiority over some supervised ML models, like support vector machine, K-nearest neighbor classifier, multilayer perceptron, long short-term memory, and statistical time series forecasting model autoregressive integrated moving average. This pre-trained stacked-BDLSTM is applied to predict the concentrations of PM_2.5 and PM₁₀ during the pandemic situation of two cases, viz., complete lockdown and partial lockdown using a deep model-based transfer learning (TL) approach (TLS-BDLSTM). Transfer learning aims to utilize the information gained from one problem to improve the predictive performance of a learning model for a different but related problem. Our work helps to demonstrate how TL is useful when there is a scarcity of data during the COVID-19 pandemic regarding the drastic change in concentration of pollutants. The results reveal the best prediction performance of TLS-BDLSTM with a lead time of 24 h as compared to some well-known traditional ML and statistical models and the pre-trained stacked-BDLSTM. The prediction is then validated using the real-time data obtained during the complete lockdown due to COVID second wave (16 May-15 June 2021) with different time steps, e.g., 24 h, 48 h, 72 h, and 96-120 h. TLS-BDLSTM involving transfer learning is seen to outperform the said comparing methods in modeling the long-term temporal dependency of multivariate time series data and boost the forecast efficiency not only in single step, but also in multiple steps. The proposed methodologies are effective, consistent, and can be used by operational organizations to utilize in monitoring and management of air quality.

Z-number-based AQI in rough set theoretic framework for interpretation of air quality for different thresholds of PM2.5 and PM10

Kolkata has a reputation for being one of the world's most polluted cities, particularly in the post-monsoon months of October, November, and December. Diwali, a Hindu festival, coincides with these months where a large number of firecrackers are set off followed by high emissions of air pollutants. As a result, the air quality index (AQI) deteriorates to "very poor" (301 ≤ AQI ≤ 400) and "poor" (201 ≤ AQI ≤ 300) categories. This situation stays for several days to a month. The present study aims to identify the thresholds for PM_2.5 and PM₁₀ that cause the AQI of Kolkata to deteriorate to "very poor" and "poor." For this purpose, we have used a rough set theory-based condition-decision support system to predict the aforementioned categories of AQI. We have developed a Z-number-based novel quantification measure of semantic information of AQI to assess the reliability of the outcomes, as generated from the condition-decision-based decision rules, during post-monsoon season. The result reveals the best possible forecast of AQI with linguistic summarization of the reliability or confidence for different threshold ranges of PM₁₀ and PM_2.5. Inverse-decision rules based on rough set theory are utilized to justify and validate the forecasts. The explainability of the condition-decision support system is demonstrated/visualized using a flow graph that maps rough-rule-based different decision paths between input and output with strength, certainty, and coverage. The investigation resulted in an advanced intelligent environmental decision support system (IEDSS) for air-quality prediction.

Pollutant specific optimal deep learning and statistical model building for air quality forecasting

Poor air quality is becoming a critical environmental concern in different countries over the last several years. Most of the air pollutants have serious consequences on human health and wellbeing. In this context, efficient forecasting of air pollutants is currently crucial to predict future events with a view to taking corrective actions and framing effective environmental policies. Although deep learning (DL) as well as statistical forecasting models are investigated in the literature, they have rarely used in air pollutant-specific optimal model building for long-term forecasting. In this paper, our aim is to develop the pollutant-specific optimal forecasting models for the phases spanning from preprocessing to model building by investigating a set of predictive techniques. In this regard, this paper presents a methodology for long-term forecasting of some important air pollutants. More specifically, a total of eight best performing models such as stacked LSTM, LSTM auto-encoder, Bi-LSTM, convLSTM, Holt-Winters, auto-regressive (AR), SARIMA, and Prophet are investigated for developing pollutant-specific optimal forecasting models. The study is carried out based on the real-world data obtained from government-run air quality monitoring units in Kolkata over a period of 4 years. The models such as Holt-Winters, Bi-LSTM, and ConvLSTM achieve high forecasting accuracy with respect to MAE and RMSE values for majority of the pollutants.

Deep Learning Approach for Assessing Air Quality During COVID-19 Lockdown in Quito

Weather Normalized Models (WNMs) are modeling methods used for assessing air contaminants under a business-as-usual (BAU) assumption. Therefore, WNMs are used to assess the impact of many events on urban pollution. Recently, different approaches have been implemented to develop WNMs and quantify the lockdown effects of COVID-19 on air quality, including Machine Learning (ML). However, more advanced methods, such as Deep Learning (DL), have never been applied for developing WNMs. In this study, we proposed WNMs based on DL algorithms, aiming to test five DL architectures and compare their performances to a recent ML approach, namely Gradient Boosting Machine (GBM). The concentrations of five air pollutants (CO, NO₂, PM_2.5, SO₂, and O₃) are studied in the city of Quito, Ecuador. The results show that Long-Short Term Memory (LSTM) and Bidirectional Recurrent Neural Network (BiRNN) outperform the other algorithms and, consequently, are recommended as appropriate WNMs to quantify the effects of the lockdowns on air pollution. Furthermore, examining the variable importance in the LSTM and BiRNN models, we identify that the most relevant temporal and meteorological features for predicting air quality are Hours (time of day), Index (1 is the first collected data and increases by one after each instance), Julian Day (day of the year), Relative Humidity, Wind Speed, and Solar Radiation. During the full lockdown, the concentration of most pollutants has decreased drastically: −48.75%, for CO, −45.76%, for SO₂, −42.17%, for PM_2.5, and −63.98%, for NO₂. The reduction of this latter gas has induced an increase of O₃ by +26.54%.

Development and application of an automated air quality forecasting system based on machine learning

As one of the most concerned issues in modern society, air quality has received extensive attentions from the public and the government, which promotes the continuous development and progress of air quality forecasting technology. In this study, an automated air quality forecasting system based on machine learning has been developed and applied for daily forecasts of six common pollutants (PM_2.5, PM_10, SO₂, NO₂, O₃, and CO) and pollution levels, which can automatically find the best "Model + Hyperparameters" without human intervention. Five machine learning models and an ensemble model (Stacked Generalization) were integrated into the system, supported by a knowledge base containing the meteorological observed data, pollutant concentrations, pollutant emissions, and model reanalysis data. Then five-year data (2015-2019) of Beijing, Shanghai, Guangzhou, Chengdu, Xi'an, Wuhan, and Changchun in China, were used as an application case to study the effectiveness of the automated forecasting system. Based on the analysis of seven evaluation criteria and pollution level forecasts, combined with the forecasting results for the next 3-days, it is found that the automated system can achieve satisfactory forecasting performance, better than most of numerical model results. This implied that the developed system unveils a good application prospect in the field of environmental meteorology.

A systematic literature review of deep learning neural network for time series air quality forecasting

Rapid progress of industrial development, urbanization and traffic has caused air quality reduction that negatively affects human health and environmental sustainability, especially among developed countries. Numerous studies on the development of air quality forecasting model using machine learning have been conducted to control air pollution. As such, there are significant numbers of reviews on the application of machine learning in air quality forecasting. Shallow architectures of machine learning exhibit several limitations and yield lower forecasting accuracy than deep learning architecture. Deep learning is a new technology in computational intelligence; thus, its application in air quality forecasting is still limited. This study aims to investigate the deep learning applications in time series air quality forecasting. Owing to this, literature search is conducted thoroughly from all scientific databases to avoid unnecessary clutter. This study summarizes and discusses different types of deep learning algorithms applied in air quality forecasting, including the theoretical backgrounds, hyperparameters, applications and limitations. Hybrid deep learning with data decomposition, optimization algorithm and spatiotemporal models are also presented to highlight those techniques' effectiveness in tackling the drawbacks of individual deep learning models. It is clearly stated that hybrid deep learning was able to forecast future air quality with higher accuracy than individual models. At the end of the study, some possible research directions are suggested for future model development. The main objective of this review study is to provide a comprehensive literature summary of deep learning applications in time series air quality forecasting that may benefit interested researchers for subsequent research.

Real-Time In-Vehicle Air Quality Monitoring System Using Machine Learning Prediction Algorithm

This paper presents the development of a real-time cloud-based in-vehicle air quality monitoring system that enables the prediction of the current and future cabin air quality. The designed system provides predictive analytics using machine learning algorithms that can measure the drivers’ drowsiness and fatigue based on the air quality presented in the cabin car. It consists of five sensors that measure the level of CO₂, particulate matter, vehicle speed, temperature, and humidity. Data from these sensors were collected in real-time from the vehicle cabin and stored in the cloud database. A predictive model using multilayer perceptron, support vector regression, and linear regression was developed to analyze the data and predict the future condition of in-vehicle air quality. The performance of these models was evaluated using the Root Mean Square Error, Mean Squared Error, Mean Absolute Error, and coefficient of determination (R²). The results showed that the support vector regression achieved excellent performance with the highest linearity between the predicted and actual data with an R² of 0.9981.

A recurrent neural network using historical data to predict time series indoor PM2.5 concentrations for residential buildings

Due to the severe outdoor PM2.5 pollution in China, many people have installed air-cleaning systems in homes. To make the systems run automatically and intelligently, we developed a recurrent neural network (RNN) that uses historical data to predict the future indoor PM2.5 concentration. The RNN architecture includes an autoencoder and a recurrent part. We used data measured in an apartment over the course of an entire year to train and test the RNN. The data include indoor/outdoor PM2.5 concentration, environmental parameters and time of day. By comparing three different input strategies, we found that a strategy employing historical PM2.5 and time of day as inputs performed best. With this strategy, the model can be applied to predict the relatively stable trend of indoor PM2.5 concentration in advance. When the input length is 2 h and the prediction horizon is 30 min, the median prediction error is 8.3 µg/m³ for the whole test set. For times with indoor PM2.5 concentrations between (20,50] µg/m³ and (50,100] µg/m³ , the median prediction error is 8.3 and 9.2 µg/m³ , respectively. The low prediction error between the ground-truth and predicted values shows that the RNN can predict indoor PM2.5 concentrations with satisfactory performance.

Hourly Ground-Level PM2.5 Estimation Using Geostationary Satellite and Reanalysis Data via Deep Learning

This study proposes an improved approach for monitoring the spatial concentrations of hourly particulate matter less than 2.5 μm in diameter (PM2.5) via a deep neural network (DNN) using geostationary ocean color imager (GOCI) images and unified model (UM) reanalysis data over the Korean Peninsula. The DNN performance was optimized to determine the appropriate training model structures, incorporating hyperparameter tuning, regularization, early stopping, and input and output variable normalization to prevent training dataset overfitting. Near-surface atmospheric information from the UM was also used as an input variable to spatially generalize the DNN model. The retrieved PM2.5 from the DNN was compared with estimates from random forest, multiple linear regression, and the Community Multiscale Air Quality model. The DNN demonstrated the highest accuracy compared to that of the conventional methods for the hold-out validation (root mean square error (RMSE) = 7.042 μg/m3, mean bias error (MBE) = −0.340 μg/m3, and coefficient of determination (R2) = 0.698) and the cross-validation (RMSE = 9.166 μg/m3, MBE = 0.293 μg/m3, and R2 = 0.49). Although the R2 was low due to underestimated high PM2.5 concentration patterns, the RMSE and MBE demonstrated reliable accuracy values (<10 μg/m3 and 1 μg/m3, respectively) for the hold-out validation and cross-validation.

Attention-based parallel networks (APNet) for PM2.5 spatiotemporal prediction

Urban particulate matter forecast is an important part of air pollution early warning and control management, especially the forecast of fine particulate matter (PM_2.5). However, the existing PM_2.5 concentration prediction methods cannot effectively capture the complex nonlinearity of PM_2.5 concentration, and most of them cannot accurately simulate the temporal and spatial dependence of PM_2.5 concentration at the same time. In this paper, we propose an attention-based parallel network (APNet), which can extract short-term and long-term temporal features simultaneously based on the attention-based CNN-LSTM multilayer structure to predict PM_2.5 concentration in the next 72 h. Firstly, the Maximum Information Coefficient (MIC) is designed for spatiotemporal correlation analysis, fully considering the linearity, non-linearity and non-functionality between the data of each monitoring station. The potential inherent features of the input data are effectively extracted through the convolutional neural network (CNN). Then, an optimized long short-term memroy (LSTM) network captures the short-term mutations of the time series. An attention mechanism is further designed for the proposed model, which automatically assigns different weights to different feature states at different time stages to distinguish their importance, and can achieve precise temporal and spatial interpretability. In order to further explore the long-term time features, we propose a Bi-LSTM parallel module to extract the periodic characteristics of PM_2.5 concentration from both previous and posterior directions. Experimental results based on a real-world dataset indicates that the proposed model outperforms other existing state-of-the-art methods. Moreover, evaluations of recall (0.790), precision (0.848) (threshold: 151 μg/m³) for 72 h prediction also verify the feasibility of our proposed model. The methodology can be used for predicting other multivariate time series data in the future.

PM2.5 concentration forecasting at surface monitoring sites using GRU neural network based on empirical mode decomposition

The main component of haze is the particulate matter (PM) 2.5. How to explore the laws of PM2.5 concentration changes is the main content of air quality prediction. Combining the characteristics of temporality and non-linearity in PM2.5 concentration series, more and more deep learning methods are currently applied to PM2.5 predictions, but most of them ignore the non-stationarity of time series, which leads to a lower accuracy of model prediction. To address this issue, an integration method of gated recurrent unit neural network based on empirical mode decomposition (EMD-GRU) for predicting PM2.5 concentration was proposed in this paper. This method uses empirical mode decomposition (EMD) to decompose the PM2.5 concentration sequence first and then fed the multiple stationary sub-sequences obtained after the decomposition and the meteorological features into the constructed GRU neural network successively for training and predicting. Finally, the sub-sequences of the prediction output are added to obtain the prediction results of PM2.5 concentration. The forecast result of the case in this paper show that the EMD-GRU model reduces the RMSE by 44%, MAE by 40.82%, and SMAPE by 11.63% compared to the single GRU model.

Long-term time-series pollution forecast using statistical and deep learning methods

Tackling air pollution has become of utmost importance since the last few decades. Different statistical as well as deep learning methods have been proposed till now, but seldom those have been used to forecast future long-term pollution trends. Forecasting long-term pollution trends into the future is highly important for government bodies around the globe as they help in the framing of efficient environmental policies. This paper presents a comparative study of various statistical and deep learning methods to forecast long-term pollution trends for the two most important categories of particulate matter (PM) which are PM2.5 and PM10. The study is based on Kolkata, a major city on the eastern side of India. The historical pollution data collected from government set-up monitoring stations in Kolkata are used to analyse the underlying patterns with the help of various time-series analysis techniques, which is then used to produce a forecast for the next two years using different statistical and deep learning methods. The findings reflect that statistical methods such as auto-regressive (AR), seasonal auto-regressive integrated moving average (SARIMA) and Holt–Winters outperform deep learning methods such as stacked, bi-directional, auto-encoder and convolution long short-term memory networks based on the limited data available.

Characteristics of HONO and its impact on O3 formation in the Seoul Metropolitan Area during the Korea-US Air Quality Study

Photolysis of nitrous acid (HONO) is recognized as an early-morning source of OH radicals in the urban air. During the Korea-US air quality (KORUS-AQ) campaign, HONO was measured using quantum cascade - tunable infrared laser differential absorption spectrometer (QC-TILDAS) at Olympic Park in Seoul from 17 May, 2016 to 14 June, 2016. The HONO concentration was in the range of 0.07-3.46 ppbv, with an average of 0.93 ppbv. Moreover, it remained high from 00:00-05:00 LST. During this time, the mean concentration was higher during the high-O3 episodes (1.82 ppbv) than the non-episodes (1.20 ppbv). In the morning, the OH radicals that were produced from HONO photolysis were 50% higher (0.95 pptv) during the high-O3 episodes than the non-episodes. Diurnal variations in HOx and O3 concentrations were simulated by the F0AM model, which revealed a difference of ~20 ppbv in the daily maximum O3 concentrations between the high-O3 episodes and non-episodes. Furthermore, the HONO concentration increased with an increase in relative humidity (RH) up to 80%; the highest HONO was associated with the top 10% NO2 in each RH group, confirming that NO2 is one of the main precursors of HONO. At night, the conversion ratio of NO2 to HONO was estimated to be 0.88×10-2 h-1; this ratio was found to increase with an increase in RH. The Aitken mode particles (30-120 nm), which act as catalyst surfaces, exhibited a similar tendency with a conversion ratio that increased along with RH, indicating the coupling of surfaces with HONO conversion. Using an artificial neural network (ANN) model, HONO concentrations were successfully simulated with measured variables (r2 = 0.66 as an average of five models). Among these variables, NOx, aerosol surface area, and RH were found to be the main factors affecting the ambient HONO concentrations. The results reveal that RH facilitates the conversion of NO2 to HONO by constraining the availability of aerosol surfaces. This study demonstrates the coupling of HONO with the HOx-O3 cycle in the Seoul Metropolitan Area (SMA) and provides practical evidence of the heterogeneous formation of HONO by employing the ANN model.

PM2.5 concentration modeling and prediction by using temperature-based deep belief network

Air quality prediction is a global hot issue, and PM_2.5 is an important factor affecting air quality. Due to complicated causes of formation, PM_2.5 prediction is a thorny and challenging task. In this paper, a novel deep learning model named temperature-based deep belief networks (TDBN) is proposed to predict the daily concentrations of PM_2.5 for the next day. Firstly, the location of PM_2.5 concentration prediction is Chaoyang Park in Beijing of China from January 1, 2018 to October 27, 2018. The auxiliary variables are selected as input variables of TDBN by Partial Least Square (PLS), and the corresponding data is divided into three independent sections: training samples, validating samples and testing samples. Secondly, the TDBN is composed of temperature-based restricted Boltzmann machine (RBM), where temperature is considered as an effective physical parameter in energy balance of training RBM. The structural parameters of TDBN are determined by minimizing the error in the training process, including hidden layers number, hidden neurons and value of temperature. Finally, the testing samples are used to test the performance of the proposed TDBN on PM_2.5 prediction, and the other similar models are tested by the same testing samples for convenience of comparison with TDBN. The experimental results demonstrate that TDBN performs better than its peers in root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R²).

An Integrated Approach of Belief Rule Base and Deep Learning to Predict Air Pollution

Sensor data are gaining increasing global attention due to the advent of Internet of Things (IoT). Reasoning is applied on such sensor data in order to compute prediction. Generating a health warning that is based on prediction of atmospheric pollution, planning timely evacuation of people from vulnerable areas with respect to prediction of natural disasters, etc., are the use cases of sensor data stream where prediction is vital to protect people and assets. Thus, prediction accuracy is of paramount importance to take preventive steps and avert any untoward situation. Uncertainties of sensor data is a severe factor which hampers prediction accuracy. Belief Rule Based Expert System (BRBES), a knowledge-driven approach, is a widely employed prediction algorithm to deal with such uncertainties based on knowledge base and inference engine. In connection with handling uncertainties, it offers higher accuracy than other such knowledge-driven techniques, e.g., fuzzy logic and Bayesian probability theory. Contrarily, Deep Learning is a data-driven technique, which constitutes a part of Artificial Intelligence (AI). By applying analytics on huge amount of data, Deep Learning learns the hidden representation of data. Thus, Deep Learning can infer prediction by reasoning over available data, such as historical data and sensor data streams. Combined application of BRBES and Deep Learning can compute prediction with improved accuracy by addressing sensor data uncertainties while utilizing its discovered data pattern. Hence, this paper proposes a novel predictive model that is based on the integrated approach of BRBES and Deep Learning. The uniqueness of this model lies in the development of a mathematical model to combine Deep Learning with BRBES and capture the nonlinear dependencies among the relevant variables. We optimized BRBES further by applying parameter and structure optimization on it. Air pollution prediction has been taken as use case of our proposed combined approach. This model has been evaluated against two different datasets. One dataset contains synthetic images with a corresponding label of PM2.5 concentrations. The other one contains real images, PM2.5 concentrations, and numerical weather data of Shanghai, China. We also distinguished a hazy image between polluted air and fog through our proposed model. Our approach has outperformed only BRBES and only Deep Learning in terms of prediction accuracy.

Urban PM2.5 Concentration Prediction via Attention-Based CNN–LSTM

Urban particulate matter forecasting is regarded as an essential issue for early warning and control management of air pollution, especially fine particulate matter (PM2.5). However, existing methods for PM2.5 concentration prediction neglect the effects of featured states at different times in the past on future PM2.5 concentration, and most fail to effectively simulate the temporal and spatial dependencies of PM2.5 concentration at the same time. With this consideration, we propose a deep learning-based method, AC-LSTM, which comprises a one-dimensional convolutional neural network (CNN), long short-term memory (LSTM) network, and attention-based network, for urban PM2.5 concentration prediction. Instead of only using air pollutant concentrations, we also add meteorological data and the PM2.5 concentrations of adjacent air quality monitoring stations as the input to our AC-LSTM. Hence, the spatiotemporal correlation and interdependence of multivariate air quality-related time-series data are learned by the CNN–LSTM network in AC-LSTM. The attention mechanism is applied to capture the importance degrees of the effects of featured states at different times in the past on future PM2.5 concentration. The attention-based layer can automatically weigh the past feature states to improve prediction accuracy. In addition, we predict the PM2.5 concentrations over the next 24 h by using air quality data in Taiyuan city, China, and compare it with six baseline methods. To compare the overall performance of each method, the mean absolute error (MAE), root-mean-square error (RMSE), and coefficient of determination (R2) are applied to the experiments in this paper. The experimental results indicate that our method is capable of dealing with PM2.5 concentration prediction with the highest performance.

Deep Learning-based Framework for Smart Sustainable Cities

Pure air is vital for sustaining human life. Air pollution causes long-term effects on people. There is an urgent need for protecting people from its profound effects. In general, people are unaware of the levels to which they are exposed to air pollutants. Vehicles, burning various kinds of waste, and industrial gases are the top three onset agents of air pollution. Of these three top agents, human beings are exposed frequently to the pollutants due to motor vehicles. To aid in protecting people from vehicular air pollutants, this article proposes a framework that utilizes deep learning models. The framework utilizes a deep belief network to predict the levels of air pollutants along the paths people travel and also a comparison with the predictions made by a feed forward neural network and an extreme learning machine. When evaluating the deep belief neural network for the case study undertaken, a deep belief network was able to achieve a higher index of agreement and lower RMSE values.

Social Big-Data Analysis of Particulate Matter, Health, and Society

The study collected particulate matter (PM)-related documents in Korea and classified main keywords related to particulate matter, health, and social problems using text and opinion mining. The study attempted to present a prediction model for important causes related to particulate matter by using social big-data analysis. Topics related to particulate matter were collected from online (online news sites, blogs, cafés, social network services, and bulletin boards) from 1 January 2015, to 31 May 2016, and 226,977 text documents were included in the analysis. The present study applied machine-learning analysis technique to forecast the risk of particulate matter. Emotions related to particulate matter were found to be 65.4% negative, 7.7% neutral, and 27.0% positive. Intelligent services that can detect early and prevent unknown crisis situations of particulate matter may be possible if risk factors of particulate matter are predicted through the linkage of the machine-learning prediction model.

Markov Chain Model Development for Forecasting Air Pollution Index of Miri, Sarawak

A Markov chain is commonly used in stock market analysis, manpower planning, and in many other areas because of its efficiency in predicting long run behavior. However, the Air Quality Index (AQI) suffers from not using a Markov chain in its forecasting approach. Therefore, this paper proposes a simple forecasting tool to predict the future air quality with a Markov chain model. The proposed method introduces the Markov chain as an operator to evaluate the distribution of the pollution level in the long term. Initial state vector and state transition probability were used in forecasting the behavior of Air Pollution Index (API) that has been obtained from the observed frequency for one state shift to another. The study explores that regardless of the present status of API, in the long run, the index shows a probability of 0.9231 for a good state, and a moderate and unhealthy state with a probability of 0.0722 and 0.0037, while for very unhealthy and hazardous states a probability of 0.0001 and 0.0009. The outcome of this study reveals that the model development could be used as a forecasting method that able to help government to project a prevention action plan during hazy weather.

A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memory

Increasing availability of data related to air quality from ground monitoring stations has provided the chance for data mining researchers to propose sophisticated models for predicting the concentrations of different air pollutants. In this paper, we proposed a hybrid model based on deep learning methods that integrates Graph Convolutional networks and Long Short-Term Memory networks (GC-LSTM) to model and forecast the spatiotemporal variation of PM_2.5 concentrations. Specifically, historical observations on different stations are constructed as spatiotemporal graph series, and historical air quality variables, meteorological factors, spatial terms and temporal attributes are defined as graph signals. To evaluate the performance of the GC-LSTM, we compared our results with several state-of-the-art methods in different time intervals. Based on the results, our GC-LSTM model achieved the best performance for predictions. Moreover, evaluations of recall rate (68.45%), false alarm rate (4.65%) (both of threshold: 115 μg/m³) and correlation coefficient R² (0.72) for 72-hour predictions also verify the feasibility of our proposed model. This methodology can be used for concentration forecasting of different air pollutants in future.

An ensemble long short-term memory neural network for hourly PM2.5 concentration forecasting

To protect public health by providing an early warning, PM_2.5 concentration forecasting is an essential and effective work. In this paper, an ensemble long short-term memory neural network (E-LSTM) is proposed for hourly PM_2.5 concentration forecasting. The presented model is implemented using three steps: (1) ensemble empirical mode decomposition (EEMD) is firstly utilized for multi-modal feature extraction, (2) long short-term memory approach (LSTM) is then employed for multi-modal feature learning, and (3) inverse EEMD computation is finally used for multi-modal feature estimated integration. In each modeling process, the mode information of the PM_2.5 and the corresponding meteorological variables in 1-h advance are utilized as inputs to forecast the next mode information of the PM_2.5 concentration. To evaluate the performance of the E-LSTM model, two datasets collected from two environmental monitoring stations in Beijing, China, are investigated. It is demonstrated that the E-LSTM model inspired by ensemble learning, which constructs multiple LSTMs in different modes, obtained better forecasting performance than that using the single LSTM and feed forward neural network in terms of mean absolute percentage error (19.604% and 16.929%), root mean square error (12.077 μg m^-3 and 13.983 μg m^-3), and correlation coefficient criteria (0.994 and 0.991) respectively.

Prediction of Air Pollution Concentration Based on mRMR and Echo State Network

Air pollution has become a global environmental problem, because it has a great adverse impact on human health and the climate. One way to explore this problem is to monitor and predict air quality index in an economical way. Accurate monitoring and prediction of air quality index (AQI), e.g., PM2.5 concentration, is a challenging task. In order to accurately predict the PM2.5 time series, we propose a supplementary leaky integrator echo state network (SLI-ESN) in this paper. It adds the historical state term of the historical moment to the calculation of leaky integrator reservoir, which improves the influence of historical evolution state on the current state. Considering the redundancy and correlation between multivariable time series, minimum redundancy maximum relevance (mRMR) feature selection method is introduced to reduce redundant and irrelevant information, and increase computation speed. A variety of evaluation indicators are used to assess the overall performance of the proposed method. The effectiveness of the proposed model is verified by the experiment of Beijing PM2.5 time series prediction. The comparison of learning time also shows the efficiency of the algorithm.

Deep Learning in the Biomedical Applications: Recent and Future Status

Deep neural networks represent, nowadays, the most effective machine learning technology in biomedical domain. In this domain, the different areas of interest concern the Omics (study of the genome—genomics—and proteins—transcriptomics, proteomics, and metabolomics), bioimaging (study of biological cell and tissue), medical imaging (study of the human organs by creating visual representations), BBMI (study of the brain and body machine interface) and public and medical health management (PmHM). This paper reviews the major deep learning concepts pertinent to such biomedical applications. Concise overviews are provided for the Omics and the BBMI. We end our analysis with a critical discussion, interpretation and relevant open challenges.

Deep learning opens new horizons in personalized medicine

Although the idea of the personalization of patient care dates back to the time of Hippocrates, recent advances in diagnostic medical imaging and molecular medicine are gradually transforming healthcare services, by offering information and diagnostic tools enabling individualized patient management. Facilitating personalized / precision medicine requires taking into account multiple heterogenous parameters, such as sociodemographics, gene variability, environmental and lifestyle factors. Therefore, one of the most critical challenges in personalized medicine is the need to transform large, multi-modal data into decision support tools, capable of bridging the translational gap to the clinical setting. Towards these challenges, deep learning (DL) provides a novel approach, which enables obtaining or developing high-accuracy, multi-modal predictive models, that allow the implementation of the personalized medicine vision in the near future. DL is a highly effective strategy in addressing these challenges, with DL-based models leading to unprecedented results, matching or even improving state-of-the-art prediction/detection rates based on both intuitive and non-intuitive disease descriptors. These results hold promise for significant socio-economic benefits from the application of DL personalized medicine.

A novel spatiotemporal convolutional long short-term neural network for air pollution prediction

Air pollution is a serious environmental problem that has drawn worldwide attention. Predicting air pollution in advance has great significance on people's daily health control and government decision-making. However, existing research methods have failed to effectively extract the spatiotemporal features of air pollutant concentration data, and exhibited low precision in long-term predictions and sudden changes in air quality. In the present study, a spatiotemporal convolutional long short-term memory neural network extended (C-LSTME) model for predicting air quality concentration was proposed. In order to encompass the spatiality and temporality of the data, the model involved the historical air pollutant concentration of the present station, as well as that of the adaptive k-nearest neighboring stations, into the model. High-level spatiotemporal features were extracted through the combination of the convolutional neural network (CNN) and long short-term memory neural network (LSTM-NN), and meteorological data and aerosol data were also integrated, in order to improve model prediction performance. Hourly PM2.5 (particulate matter with an aerodynamic diameter of ≤2.5 mm) concentration data collected at 1233 air quality monitoring stations in Beijing and the whole China from January 1, 2016 to December 31, 2017 were used to validate the effectiveness of the proposed C-LSTME model. The results show that the present model has achieved better performance than current state-of-the-art models for different time predictions at different regional scales.

Multimodal MRI-Based Triage for Acute Stroke Therapy: Challenges and Progress

Revascularization therapies have been established as the treatment mainstay for acute ischemic stroke. However, a substantial number of patients are either ineligible for revascularization therapy, or the treatment fails or is futile. At present, non-contrast computed tomography is the first-line neuroimaging modality for patients with acute stroke. The use of magnetic resonance imaging (MRI) to predict the response to early revascularization therapy and to identify patients for delayed treatment is desirable. MRI could provide information on stroke pathophysiologies, including the ischemic core, perfusion, collaterals, clot, and blood–brain barrier status. During the past 20 years, there have been significant advances in neuroimaging as well as in revascularization strategies for treating patients with acute ischemic stroke. In this review, we discuss the role of MRI and post-processing, including machine-learning techniques, and recent advances in MRI-based triage for revascularization therapies in acute ischemic stroke.

Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation

Air pollutant concentration forecasting is an effective method of protecting public health by providing an early warning against harmful air pollutants. However, existing methods of air pollutant concentration prediction fail to effectively model long-term dependencies, and most neglect spatial correlations. In this paper, a novel long short-term memory neural network extended (LSTME) model that inherently considers spatiotemporal correlations is proposed for air pollutant concentration prediction. Long short-term memory (LSTM) layers were used to automatically extract inherent useful features from historical air pollutant data, and auxiliary data, including meteorological data and time stamp data, were merged into the proposed model to enhance the performance. Hourly PM_2.5 (particulate matter with an aerodynamic diameter less than or equal to 2.5 μm) concentration data collected at 12 air quality monitoring stations in Beijing City from Jan/01/2014 to May/28/2016 were used to validate the effectiveness of the proposed LSTME model. Experiments were performed using the spatiotemporal deep learning (STDL) model, the time delay neural network (TDNN) model, the autoregressive moving average (ARMA) model, the support vector regression (SVR) model, and the traditional LSTM NN model, and a comparison of the results demonstrated that the LSTME model is superior to the other statistics-based models. Additionally, the use of auxiliary data improved model performance. For the one-hour prediction tasks, the proposed model performed well and exhibited a mean absolute percentage error (MAPE) of 11.93%. In addition, we conducted multiscale predictions over different time spans and achieved satisfactory performance, even for 13-24 h prediction tasks (MAPE = 31.47%).