Prediction model for air particulate matter levels in the households of elderly individuals in Hong Kong

Assessing household fine particulate matter (PM2.5) through measurement and modeling in the Bangladesh cook stove pregnancy cohort study (CSPCS)

Biomass fuel burning is a significant contributor of household fine particulate matter (PM2.5) in the low to middle income countries (LMIC) and assessing PM2.5 levels is essential to investigate exposure-related health effects such as pregnancy outcomes and acute lower respiratory infection in infants. However, measuring household PM2.5 requires significant investments of labor, resources, and time, which limits the ability to conduct health effects studies. It is therefore imperative to leverage lower-cost measurement techniques to develop exposure models coupled with survey information about housing characteristics. Between April 2017 and March 2018, we continuously sampled PM2.5 in three seasonal waves for approximately 48-h (range 46 to 52-h) in 74 rural and semi-urban households among the participants of the Bangladesh Cook Stove Pregnancy Cohort Study (CSPCS). Measurements were taken simultaneously in the kitchen, bedroom, and open space within the household. Structured questionnaires captured household-level information related to the sources of air pollution. With data from two waves, we fit multivariate mixed effect models to estimate 24-h average, cooking time average, daytime and nighttime average PM2.5 in each of the household locations. Households using biomass cookstoves had significantly higher PM2.5 concentrations than those using electricity/liquefied petroleum gas (626 μg/m3 vs. 213 μg/m3). Exposure model performances showed 10-fold cross validated R2 ranging from 0.52 to 0.76 with excellent agreement in independent tests against measured PM2.5 from the third wave of monitoring and ambient PM2.5 from a separate satellite-based model (correlation coefficient, r = 0.82). Significant predictors of household PM2.5 included ambient PM2.5, season, and types of fuel used for cooking. This study demonstrates that we can predict household PM2.5 with moderate to high confidence using ambient PM2.5 and household characteristics. Our results present a framework for estimating household PM2.5 exposures in LMICs, which are often understudied and underrepresented due to resource limitations.

Construction and evaluation of hourly average indoor PM2.5 concentration prediction models based on multiple types of places

Background

People usually spend most of their time indoors, so indoor fine particulate matter (PM2.5) concentrations are crucial for refining individual PM2.5 exposure evaluation. The development of indoor PM2.5 concentration prediction models is essential for the health risk assessment of PM2.5 in epidemiological studies involving large populations.

Methods

In this study, based on the monitoring data of multiple types of places, the classical multiple linear regression (MLR) method and random forest regression (RFR) algorithm of machine learning were used to develop hourly average indoor PM2.5 concentration prediction models. Indoor PM2.5 concentration data, which included 11,712 records from five types of places, were obtained by on-site monitoring. Moreover, the potential predictor variable data were derived from outdoor monitoring stations and meteorological databases. A ten-fold cross-validation was conducted to examine the performance of all proposed models.

Results

The final predictor variables incorporated in the MLR model were outdoor PM2.5 concentration, type of place, season, wind direction, surface wind speed, hour, precipitation, air pressure, and relative humidity. The ten-fold cross-validation results indicated that both models constructed had good predictive performance, with the determination coefficients (R2) of RFR and MLR were 72.20 and 60.35%, respectively. Generally, the RFR model had better predictive performance than the MLR model (RFR model developed using the same predictor variables as the MLR model, R2 = 71.86%). In terms of predictors, the importance results of predictor variables for both types of models suggested that outdoor PM2.5 concentration, type of place, season, hour, wind direction, and surface wind speed were the most important predictor variables.

Conclusion

In this research, hourly average indoor PM2.5 concentration prediction models based on multiple types of places were developed for the first time. Both the MLR and RFR models based on easily accessible indicators displayed promising predictive performance, in which the machine learning domain RFR model outperformed the classical MLR model, and this result suggests the potential application of RFR algorithms for indoor air pollutant concentration prediction.

Quantifying the dynamic characteristics of indoor air pollution using real-time sensors: Current status and future implication

People generally spend most of their time indoors, making indoor air quality be of great significance to human health. Large spatiotemporal heterogeneity of indoor air pollution can be hardly captured by conventional filter-based monitoring but real-time monitoring. Real-time monitoring is conducive to change air assessment mode from static and sparse analysis to dynamic and massive analysis, and has made remarkable strides in indoor air evaluation. In this review, the state of art, strengths, challenges, and further development of real-time sensors used in indoor air evaluation are focused on. Researches using real-time sensors for indoor air evaluation have increased rapidly since 2018, and are mainly conducted in China and the USA, with the most frequently investigated air pollutants of PM_2.5. In addition to high spatiotemporal resolution, real-time sensors for indoor air evaluation have prominent advantages in 3-dimensional monitoring, pollution peak and source identification, and short-term health effect evaluation. Huge amounts of data from real-time sensors also facilitate the modeling and prediction of indoor air pollution. However, challenges still remain in extensive deployment of real-time sensors indoors, including the selection, performance, stability, as well as calibration of sensors. In future, sensors with high performance, long-term stability, low price, and low energy consumption are welcomed. Furthermore, more target air pollutants are also expected to be detected simultaneously by real-time sensors in indoor air monitoring.

Chronic exposure to indoor air pollutants in association with attention-deficit/hyperactivity disorder symptoms in Chinese schoolchildren: A cross-sectional study

BACKGROUND

While increasing studies confirmed the adverse effects of indoor air pollution (IAP) on cardiopulmonary systems, less is known about the impact of IAP on child's brain.

OBJECTIVE

To explore the associations between multiple indoor air pollutants exposures and attention-deficit/hyperactivity disorder (ADHD) symptoms in Chinese schoolchildren.

METHODS

We invited 8630 individuals aged 6-12 years from an ongoing school-based cohort study across Guangzhou from April to May 2019. There are 7495 and 7245 children were respectively evaluated on the parent- and teacher-rated Conner's Rating Scale-Revised, and 7087 children were assessed on both versions. Indoor air pollutants exposures including cooking oil fumes, incense burning, home renovation, and secondhand smoke, were measured using a questionnaire reported by parents and children, and further converted into an index. Generalized linear mixed-effects models were performed to evaluate the associations between indoor air pollutants exposures and ADHD index and the presence of ADHD symptoms.

RESULTS

As reported by parents, 321 (4.3%) children had ADHD symptoms. Each of the four pollutants was positively associated with higher ADHD index and higher odds of ADHD symptoms. Children exposed to 1, 2, and ≥ 3 types of indoor air pollutants had higher ADHD index and higher odds of ADHD symptoms than those non-exposed children. For parent-reported ADHD symptoms, the odds ratios ranged from 1.24 [95% confidence interval (CI): 0.92-1.67] to 2.73 (95% CI: 1.86-4.01). These associations were consistent in parent- and teacher-reported ADHD symptoms, and the combination of both.

CONCLUSION

Indoor air pollutants exposures were positively associated with higher prevalence of children's ADHD symptoms assessed by whether parents or teachers.

Ranking the environmental factors of indoor air quality of metropolitan independent coffee shops by Random Forests model

Independent coffee shops are the alternative workplaces for people working remotely from traditional offices but are not concerned about their indoor air quality (IAQ). This study aimed to rank the environmental factors in affecting the IAQ by Random Forests (RFs) models. The indoor environments and human activities of participated independent coffee shops were observed and recorded for 3 consecutive days including weekdays and weekend during the business hours. The multi-sized particulate matter (PM), particle-bound polycyclic aromatic hydrocarbons (p-PAHs), total volatile organic compounds (TVOCs), CO, CO₂, temperature and relative humidity were monitored. RFs models ranked the environmental factors. More than 20% of the 15-min average concentrations of PM₁₀, PM_2.5, and CO₂ exceeded the World Health Organization guidelines. Occupant density affected TVOCs, p-PAHs and CO₂ concentrations directly. Tobacco smoking dominated PM₁₀, PM_2.5, TVOCs and p-PAHs concentrations mostly. CO concentration was affected by roasting bean first and tobacco smoking secondly. The non-linear relationships between temperature and these pollutants illustrated the relative low concentrations happened at temperature between 22 and 24 °C. Tobacco smoking, roasting beans and occupant density are the observable activities to alert the IAQ change. Decreasing CO₂ and optimizing the room temperature could also be the surrogate parameters to assure the IAQ.

ADFIST: Adaptive Dynamic Fuzzy Inference System Tree Driven by Optimized Knowledge Base for Indoor Air Quality Assessment

Air quality levels do not just affect climate change; rather, it leaves a significant impact on public health and wellbeing. Indoor air pollution is the major contributor to increased mortality and morbidity rates. This paper is focused on the assessment of indoor air quality based on several important pollutants (PM₁₀, PM_2.5, CO₂, CO, tVOC, and NO₂). These pollutants are responsible for potential health issues, including respiratory disease, central nervous system dysfunction, cardiovascular disease, and cancer. The pollutant concentrations were measured from a rural site in India using an Internet of Things-based sensor system. An Adaptive Dynamic Fuzzy Inference System Tree was implemented to process the field variables. The knowledge base for the proposed model was designed using a global optimization algorithm. However, the model was tuned using a local search algorithm to achieve enhanced prediction performance. The proposed model gives normalized root mean square error of 0.6679, 0.6218, 0.1077, 0.2585, 0.0667 and 0.0635 for PM₁₀, PM_2.5, CO₂, CO, tVOC, and NO₂, respectively. This approach was compared with the existing studies in the literature, and the approach was also validated against the online benchmark dataset.

Contributions of internal emissions to peaks and incremental indoor PM2.5 in rural coal use households

Indoor air quality is critically important to the human as people spend most time indoors. Indoor PM_2.5 is related to the outdoor levels, but more directly influenced by internal sources. Severe household air pollution from solid fuel use has been recognized as one major risk for human health especailly in rural area, however, the issue is significantly overlooked in most national air quality controls and intervention policies. Here, by using low-cost sensors, indoor PM_2.5 in rural homes burning coals was monitored for ~4 months and analyzed for its temporal dynamics, distributions, relationship with outdoor PM_2.5, and quantitative contributions of internal sources. A bimodal distribution of indoor PM_2.5 was identified and the bimodal characteristic was more significant at the finer time resolution. The bimodal distribution maxima were corresponding to the emissions from strong internal sources and the influence of outdoor PM_2.5, respectively. Indoor PM_2.5 was found to be correlated with the outdoor PM_2.5, even though indoor coal combustion for heating was thought to be predominant source of indoor PM_2.5. The indoor-outdoor relationship differed significantly between the heating and non-heating seasons. Impacts of typical indoor sources like cooking, heating associated with coal use, and smoking were quantitatively analyzed based on the highly time-resolved PM_2.5. Estimated contribution of outdoor PM_2.5 to the indoor PM_2.5 was ~48% during the non-heating period, but decreased to about 32% during the heating period. The contribution of indoor heating burning coals comprised up to 47% of the indoor PM_2.5 during the heating period, while the other indoor sources contributed to ~20%. The study, based on a relatively long-term timely resolved PM_2.5 data from a large number of rural households, provided informative results on temporal dynamics of indoor PM_2.5 and quantitative contributions of internal sources, promoting scientific understanding on sources and impacts of household air pollution.

Simulation and Analysis of Indoor Air Quality in Florida Using Time Series Regression (TSR) and Artificial Neural Networks (ANN) Models

Exposures to air pollutants have been associated with various acute respiratory diseases and detrimental human health. Analysis and further interpretation of air pollutant patterns are correspondingly important as monitoring them. In the present study, the 24-h and four-month indoor and outdoor PM2.5, PM10, NO2, relative humidity, and temperature were measured simultaneously for a laboratory in Gainesville city, Florida. The indoor PM2.5, PM10, and NO2 concentrations were predicted using multiple linear regression (MLR), time series regression (TSR), and artificial neural networks (ANN) models. The modeling conducted in this study aims to perform a cross comparison study between these models in a symmetric environment. The value of root-mean-square error was improved by 18.33% in comparison with the MLR model. In addition, the value of the coefficient of determination was improved by 24.68%. The ANN model had the best performance and could predict the target air pollutants at 10-min intervals of the studied building with 90% accuracy levels. The TSR model showed slightly better performance compared to the MLR model. These results can be accordingly referred for studies analyzing indoor air quality in similar building types and climate zones.

Temporal and spatial variation of PM2.5 in indoor air monitored by low-cost sensors

Indoor air pollution has significant adverse health impacts, but its spatiotemporal variations and source contributions are not well quantified. In this study, we used low-cost sensors to measure PM_2.5 concentrations in a typical apartment in Beijing. The measurements were conducted at 15 indoor sites and one outdoor site on 1-minute temporal resolution (convert to 10-minute averages for data analysis) from March 14 to 24, 2020. Based on these highly spatially-and temporally-resolved data, we characterized spatiotemporal variations and source contributions of indoor PM_2.5 in this apartment. It was found that indoor particulate matter predominantly originates from outdoor infiltration and cooking emissions with the latter contributing more fine particles. Indoor PM_2.5 concentrations were found to be correlated with ambient levels but were generally lower than those outdoors with an average I/O of 0.85. The predominant indoor source was cooking, leading to occasional high spikes. The variations observed in most rooms lagged behind those measured outdoors and in the studied kitchen. Differences between rooms were found to depend on pathway distances from sources. On average, outdoor sources contributed 36% of indoor PM_2.5, varying extensively over time and among rooms. From observed PM_2.5 concentrations at the indoor sites, source strengths, and pathway distances, a multivariate regression model was developed to predict spatiotemporal variations of PM_2.5. The model explains 79% of the observed variation and can be used to dynamically simulate PM_2.5 concentrations at any site indoors. The model's simplicity suggests the potential for regional-scale application for indoor air quality modeling.

Low-Cost Air Quality Sensing towards Smart Homes

The evolution of low-cost sensors (LCSs) has made the spatio-temporal mapping of indoor air quality (IAQ) possible in real-time but the availability of a diverse set of LCSs make their selection challenging. Converting individual sensors into a sensing network requires the knowledge of diverse research disciplines, which we aim to bring together by making IAQ an advanced feature of smart homes. The aim of this review is to discuss the advanced home automation technologies for the monitoring and control of IAQ through networked air pollution LCSs. The key steps that can allow transforming conventional homes into smart homes are sensor selection, deployment strategies, data processing, and development of predictive models. A detailed synthesis of air pollution LCSs allowed us to summarise their advantages and drawbacks for spatio-temporal mapping of IAQ. We concluded that the performance evaluation of LCSs under controlled laboratory conditions prior to deployment is recommended for quality assurance/control (QA/QC), however, routine calibration or implementing statistical techniques during operational times, especially during long-term monitoring, is required for a network of sensors. The deployment height of sensors could vary purposefully as per location and exposure height of the occupants inside home environments for a spatio-temporal mapping. Appropriate data processing tools are needed to handle a huge amount of multivariate data to automate pre-/post-processing tasks, leading to more scalable, reliable and adaptable solutions. The review also showed the potential of using machine learning technique for predicting spatio-temporal IAQ in LCS networked-systems.

A practical framework for predicting residential indoor PM2.5 concentration using land-use regression and machine learning methods

People typically spend most of their time indoors. It is of importance to establish prediction models to estimate PM_2.5 concentration in indoor environments (e.g., residential households) to allow accurate assessments of exposure in epidemiological studies. This study aimed to develop models to predict PM_2.5 concentration in residential households. PM_2.5 concentration and related parameters (e.g., basic information about the households and ventilation settings) were collected in 116 households during the winter and summer seasons in Hong Kong. Outdoor PM_2.5 concentration at households was estimated using a land-use regression model. The random forest machine learning algorithm was then applied to develop indoor PM_2.5 prediction models. The results show that the random forest model achieved a promising predictive accuracy, with R² and cross-validation R² values of 0.93 and 0.65, respectively. Outdoor PM_2.5 concentration was the most important predictor variable, followed in descending order by the household marked number, outdoor temperature, outdoor relative humidity, average household area and air conditioning. The external validation result using an independent dataset confirmed the potential application of the random forest model, with an R² value of 0.47. Overall, this study shows the value of a combined land-use regression and machine learning approach in establishing indoor PM_2.5 prediction models that provide a relatively accurate assessment of exposure for use in epidemiological studies.

Another invisible enemy indoors: COVID-19, human health, the home, and United States indoor air policy

After the emergence of the respiratory virus SARS-CoV-2 (COVID-19), many exposure and environmental health scientists promptly recognized the potentially catastrophic public health ramifications of concurrent infectious and air pollution-mediated disease. Nevertheless, much of this attention has been focused on outdoor interactions. Each year, 3.8 million people worldwide prematurely die from illnesses attributable to indoor air. Hence, poor household indoor air quality is a long-standing public health issue with even greater relevance now that many individuals are spending more time at home. At present, the Environmental Protection Agency does not regulate indoor air, and state-level legislation has resulted in a patchwork of national coverage. Here, we describe common sources of indoor air pollution, the health impacts of indoor pollutants, and populations disparately impacted by COVID-19 and poor indoor air quality. Furthermore, we detail the need for better legislation that promotes the integrity of the indoor air environment, and what individuals can do to personally protect themselves as we await more comprehensive indoor air legislation.

Early-Life Environmental Exposures and Childhood Obesity: An Exposome-Wide Approach

BACKGROUND

Chemical and nonchemical environmental exposures are increasingly suspected to influence the development of obesity, especially during early life, but studies mostly consider single exposure groups.

OBJECTIVES

Our study aimed to systematically assess the association between a wide array of early-life environmental exposures and childhood obesity, using an exposome-wide approach.

METHODS

The HELIX (Human Early Life Exposome) study measured child body mass index (BMI), waist circumference, skinfold thickness, and body fat mass in 1,301 children from six European birth cohorts age 6-11 y. We estimated 77 prenatal exposures and 96 childhood exposures (cross-sectionally), including indoor and outdoor air pollutants, built environment, green spaces, tobacco smoking, and biomarkers of chemical pollutants (persistent organic pollutants, metals, phthalates, phenols, and pesticides). We used an exposure-wide association study (ExWAS) to screen all exposure-outcome associations independently and used the deletion-substitution-addition (DSA) variable selection algorithm to build a final multiexposure model.

RESULTS

The prevalence of overweight and obesity combined was 28.8%. Maternal smoking was the only prenatal exposure variable associated with higher child BMI (z-score increase of 0.28, 95% confidence interval: 0.09, 0.48, for active vs. no smoking). For childhood exposures, the multiexposure model identified particulate and nitrogen dioxide air pollution inside the home, urine cotinine levels indicative of secondhand smoke exposure, and residence in more densely populated areas and in areas with fewer facilities to be associated with increased child BMI. Child blood levels of copper and cesium were associated with higher BMI, and levels of organochlorine pollutants, cobalt, and molybdenum were associated with lower BMI. Similar results were found for the other adiposity outcomes.

DISCUSSION

This first comprehensive and systematic analysis of many suspected environmental obesogens strengthens evidence for an association of smoking, air pollution exposure, and characteristics of the built environment with childhood obesity risk. Cross-sectional biomarker results may suffer from reverse causality bias, whereby obesity status influenced the biomarker concentration. https://doi.org/10.1289/EHP5975.