Explainable geospatial-artificial intelligence models for the estimation of PM2.5 concentration variation during commuting rush hours in Taiwan.

PM2.5 concentrations are higher during rush hours at background stations compared to the average concentration across these stations. Few studies have investigated PM2.5 concentration and its spatial distribution during rush hours using machine learning models. This study employs a geospatial-artificial intelligence (Geo-AI) prediction model to estimate the spatial and temporal variations of PM2.5 concentrations during morning and dusk rush hours in Taiwan. Mean hourly PM2.5 measurements were collected from 2006 to 2020, and aggregated into morning (7 a.m.-9 a.m.) and dusk (4 p.m.-6 p.m.) rush-hour mean concentrations. The Geo-AI prediction model was generated by integrating kriging interpolation, land-use regression, machine learning, and a stacking ensemble approach. A forward stepwise variable selection method based on the SHapley Additive exPlanations (SHAP) index was used to identify the most influential variables. The performance of the Geo-AI models for morning and dusk rush hours had accuracy scores of 0.95 and 0.93, respectively and these results were validated, indicating robust model performance. Spatially, PM2.5 concentrations were higher in southwestern Taiwan for morning rush hours, and suburban areas for dusk rush hours. Key predictors included kriged PM2.5 values, SO2 concentrations, forest density, and the distance to incinerators for both morning and dusk rush hours. These PM2.5 estimates for morning and dusk rush hours can support the development of alternative commuting routes with lower concentrations.

Anti-influenza A (H1N1) virus effect of gallic acid through inhibition of virulent protein production and association with autophagy.

Influenza remains one of the most serious infectious diseases. Gallic acid is one of the most common and representative phenolic acids found in various plants. This is an interesting subject to explore how gallic acid could inhibit H1N1 influenza virus infection by reducing the production of virulent proteins and interrupting autophagy machinery for influenza virus replication on the host cell. Cellular viability was assessed by XTT assay. The inhibitory effects on the H1N1 influenza virus were assessed by hemagglutination assay, plaque assay, and qRT-PCR. Western blot analysis was used for detecting protein levels of M1, M2, NP, LC3B, and beclin-1. Autophagy activity was demonstrated by acridine orange staining assay. The result demonstrated that there was no cytotoxic effect of gallic acid on A549 cells, and gallic acid could restore the cellular viability of H1N1 influenza virus-infected A549 cells within the experimental concentration treatment. Moreover, gallic acid could effectively restrain viral activity of the H1N1 influenza virus. After the treatment of gallic acid, the production of virulent H1N1 influenza virus proteins, that is, M1, M2, and NP protein were reduced. As for autophagic mechanism, both of the LC3B II conversion and the level ratio of LC3B II to LC3B I were notably decreased. The acridine orange staining assay also revealed decreased accumulation of autophagosomes in H1N1 influenza virus-infected cells. In conclusion, gallic acid suppresses H1N1 influenza viral infectivity through restoration of autophagy pathway and inhibition of virulent M1, M2, and NP protein production.

An ensemble mixed spatial model in estimating long-term and diurnal variations of PM2.5 in Taiwan.

Meteorology, human activities, and other emission sources drive diurnal cyclic patterns of air pollution. Previous studies mainly focused on the variation of PM2.5 concentrations during daytime rather than nighttime. In addition, assessing the spatial variations of PM2.5 in large areas is a critical issue for environmental epidemiological studies to clarify the health effects from PM2.5 exposures. In terms of air pollution spatial modelling, using only a single model might lose information in capturing spatial and temporal correlation between predictors and pollutant levels. Hence, this study aimed to propose an ensemble mixed spatial model that incorporated Kriging interpolation, land-use regression (LUR), machine learning, and stacking ensemble approach to estimate long-term PM2.5 variations for nearly three decades in daytime and nighttime. Three steps of model development were applied: 1) linear based LUR and Hybrid Kriging-LUR were used to determine influential predictors; 2) machine learning algorithms were used to enhance model prediction accuracy; 3) predictions from the selected machine learning models were fitted and evaluated again to build the final ensemble mixed spatial model. The results showed that prediction performance increased from 0.514 to 0.895 for daily, 0.478 to 0.879 for daytime, and 0.523 to 0.878 for nighttime when applying the proposed ensemble mixed spatial model compared with LUR. Results of overfitting test and extrapolation ability test confirmed the robustness and reliability of the developed models. The distance to the nearest thermal power plant, density of soil and pebbles fields, and funeral facilities might affect the variation of PM2.5 levels between daytime and nighttime. The PM2.5 level was higher in daytime compared with nighttime with little difference, revealing the importance of estimating nighttime PM2.5 variations. Our findings also clarified the emission sources in daytime and nighttime, which serve as valuable information for air pollution control strategies establishment.

Novel Aporphine- and Proaporphine-Clerodane Hybrids Identified from the Barks of Taiwanese Polyalthia longifolia (Sonn.) Thwaites var. pendula with Strong Anti-DENV2 Activity.

Hybrid natural products produced via mixed biosynthetic pathways are unique and often surprise one with unexpected medicinal properties in addition to their fascinating structural complexity/diversity. In view of chemical structures, hybridization is a way of diversifying natural products usually through dimerization of two similar or dissimilar subcomponents through a C-C or N-C covalent linkage. Here, we report four structurally attractive diterpene-alkaloid conjugates polyalongarins A-D (1-4), clerodane-containing aporphine and proaporphine alkaloids, the first of its kind from the barks of Taiwanese Polyalthia longifolia (Sonn.) Thwaites var. pendula. In addition to conventional spectroscopic analysis, single crystal X-ray crystallography was employed to determine the chemical structures and stereo-configurations of 1. Compounds 1-4 were subsequently subjected to in vitro antiviral examination against DENV2 by evaluating the expression level of the NS2B protein in DENV2-infected Huh-7 cells. These compounds display encouraging anti-DENV2 activity with superb EC50 (2.8-6.4 µM) and CC50 values (50.4-200 µM). The inhibitory mechanism of 1-4 on NS2B was further explored drawing on in-silico molecular docking analysis. Based on calculated binding affinities and predicted interactions between the functional groups of 1-4 and the allosteric-site residues of the DENV2 NS2B-NS3 protease, our analysis concludes that the clerodane-aporphine/proaporphine-type hybrids are novel and effective DENV NS2B-NS3 protease inhibitors.

Impact of mold growth on di(2-ethylhexyl) phthalate emission from moist wallpaper.

Flood damage can increase indoor concentrations of di(2-ethylhexyl) phthalate (DEHP) and molds in households with wallpaper. Wallpaper water content can affect its DEHP emission into indoor environments; however, the influence of mold growth on this DEHP emission remains unclear. Here, we evaluated whether mold growth affects DEHP emission from moist wallpaper (moist WP). Experiments were conducted in glass chambers with wallpaper containing 12.7% (w/w) DEHP and a dust tray sample system at approximately 28 °C and 100% relative humidity (RH). The experimental groups were (1) moist WP, (2) moist WP + Aspergillus versicolor (AV), (3) moist WP + Cladosporium cladosporioides, (4) moist WP + Penicillium chrysogenum, and (5) moist WP + mold mixture. Mold growth on the wallpaper and DEHP emission into air and onto dust were analyzed at nine time-points over 30 days. Initially, the moist WP group emitted relatively high concentrations of DEHP into air, but after at least 8 days, the concentration of DEHP emitted by the mold-added groups exceeded that of the moist WP group. DEHP emission onto dust, especially from the moist WP group, increased considerably at day 15. During the experimental period, the moist WP (13.63 ± 4.67 µg) and moist WP + AV (13.93 ± 0.49 µg) groups emitted higher cumulative amounts of DEHP onto dust. During the 30-day experimental period, obvious mold growth occurred over days 15-30. Moreover, the moist WP group exhibited relatively higher and lower cumulative DEHP emission into air than the mold-added groups during days 2-10 (2.71 vs. 1.94-2.94 µg) and 15-30 (1.16 vs. 1.61-2.12), respectively; a contrasting trend was observed for cumulative DEHP emission onto dust. In conclusion, mold growth affects DEHP emission from water-damaged wallpaper, and the removal or cleaning of wet wallpaper, particularly those with visible mold growth, is critical from a public health perspective.

Exploring the Potential Relationship Between Global Greenness and DALY Loss Due to Depressive Disorders.

Objective: Prior studies have shown that greenness can reduce the burden of depressive disorders. However, most were focused on local-scale analyses while limited evaluated globally. We aimed to investigate the association between greenness and the burden of depressive disorders using data from 183 countries worldwide. Methods: We used the normalized difference vegetation index (NDVI) to estimate greenness. Country-level disability-adjusted life year (DALY) loss due to depressive disorders was used to represent depressive disorder burdens. A generalized linear mixed model was applied to assess the relationship between greenness and depressive disorders after controlling for covariates. Stratified analyses were conducted to determine the effects of greenness across several socio-demographic levels. Results: The findings showed a significant negative association between greenness and the health burden of depressive disorders with a coefficient of -0.196 (95% CI: -0.356, -0.035) in the DALY changes per interquartile unit increment of NDVI. The stratified analyses suggested beneficial effects of greenness on depressive disorders across sex, various age groups especially for those aged <49 years, with low-income and/or those living in highly urbanized countries. Conclusions: Our study noted that greenness exposure was significant negative association with the burden of depressive disorders. The findings should be viewed as recommendations for relevant authorities in supporting environmental greenness enhancement to reduce the mental burdens.

Examining the Benefits of Greenness on Reducing Suicide Mortality Rate: A Global Ecological Study.

Objective: This study applied an ecological-based analysis aimed to evaluate on a global scale the association between greenness exposure and suicide mortality. Methods: Suicide mortality data provided by the Institute for Health Metrics and Evaluation and the Normalized Difference Vegetation Index (NDVI) were employed. The generalized additive mixed model was applied to evaluate with an adjustment of covariates the association between greenness and suicide mortality. Sensitivity tests and positive-negative controls also were used to examine less overt insights. Subgroup analyses were then conducted to investigate the effects of greenness on suicide mortality among various conditions. Results: The main finding of this study indicates a negative association between greenness exposure and suicide mortality, as greenness significantly decreases the risk of suicide mortality per interquartile unit increment of NDVI (relative risk = 0.69, 95%CI: 0.59-0.81). Further, sensitivity analyses confirmed the robustness of the findings. Subgroup analyses also showed a significant negative association between greenness and suicide mortality for various stratified factors, such as sex, various income levels, urbanization levels, etc. Conclusions: Greenness exposure may contribute to a reduction in suicide mortality. It is recommended that policymakers and communities increase environmental greenness in order to mitigate the global health burden of suicide.

Estimations of infiltration factors of diurnal PM2.5 and heavy metals in children's bedrooms.

Children are the sensitive population to fine particulate matter (PM2.5 ) exposure and spend most of their time in bedroom. Infiltration factor (Finf ) can be used to calculate the fraction of total indoor PM2.5 with outdoor origin to increase the accuracy of exposure assessment. However, studies have ignored the diurnal variations of PM2.5 Finf values, and a few studies have estimated Finf values for heavy metals in PM2.5 in children's bedrooms. To calculate the PM2.5 Finf , real-time indoor and outdoor PM2.5 concentrations and occupants' activities were collected in 56 study bedrooms. At 22 of the 56 study bedrooms, PM2.5 samples were also collected for heavy metals analysis. We noted the PM2.5 Finf was higher during the daytime (0.70 ± 0.23) than nighttime (0.54 ± 0.27) during the hot season, and the time of air conditioner use was longer at nighttime. The largest Finf value of heavy metal was V (0.88 ± 0.25), followed by Pb (0.85 ± 0.28), Mn (0.72 ± 0.26), Cr (0.69 ± 0.35), and Zn (0.61 ± 0.32), with a larger variation. Our findings suggest that the estimations of diurnal PM2.5 and heavy metals Finf values are necessary to increase the accuracy of exposure assessment.

Stable C and N isotopes of PM2.5 and size-segregated particles emitted from incense stick and cigarette burning.

This work measured the δ13C and δ15N signatures in PM2.5 and size-segregated particles emitted from incense stick and cigarette burning in different brands or nicotine contents for pollution source identification indoors. Three popular brands of incense stick and cigarette were selected for experiments. A personal environmental monitoring sampler and a Sioutas cascade impactor were used to collect PM2.5 and size-segregated particles, respectively, for isotopic signatures analyses. Our data showed that both δ13C and δ15N values were heavier from incense stick burning (δ13C: 27.3 ± 0.5; δ15N: 8.63 ± 1.35) than cigarette (δ13C: 28.5 ± 0.2; δ15N: 4.15 ± 0.69). The scatter plots of δ13C and TC/PM2.5 and of δ15N and TN/PM2.5 can be applied to distinguish particle pollution sources and assess the influence of cigarette burning to PM2.5 according to different nicotine contents. The δ13C values in size-segregated particles were similar to incense stick or cigarette burning; the δ13C values in PM2.5 were significantly higher than those in size-segregated particles. However, the nitrogen amount was too low in most of the size-segregated particles to analyze δ15N from incense stick and cigarette burning. These results suggest that the δ13C signatures on PM2.5 cannot represent the isotopic characteristics of size-segregated particles and δ15N has limitation for pollution source identification of different particle sizes.

Influence of Indoor Temperature Exposure on Emergency Department Visits Due to Infectious and Non-Infectious Respiratory Diseases for Older People.

Previous studies have demonstrated that outdoor temperature exposure was an important risk factor for respiratory diseases. However, no study investigates the effect of indoor temperature exposure on respiratory diseases and further assesses cumulative effect. The objective of this study is to study the cumulative effect of indoor temperature exposure on emergency department visits due to infectious (IRD) and non-infectious (NIRD) respiratory diseases among older adults. Subjects were collected from the Longitudinal Health Insurance Database in Taiwan. The cumulative degree hours (CDHs) was used to assess the cumulative effect of indoor temperature exposure. A distributed lag nonlinear model with quasi-Poisson function was used to analyze the association between CDHs and emergency department visits due to IRD and NIRD. For IRD, there was a significant risk at 27, 28, 29, 30, and 31 °C when the CDHs exceeded 69, 40, 14, 5, and 1 during the cooling season (May to October), respectively, and at 19, 20, 21, 22, and 23 °C when the CDHs exceeded 8, 1, 1, 35, and 62 during the heating season (November to April), respectively. For NIRD, there was a significant risk at 19, 20, 21, 22, and 23 °C when the CDHs exceeded 1, 1, 16, 36, and 52 during the heating season, respectively; the CDHs at 1 was only associated with the NIRD at 31 °C during the cooling season. Our data also indicated that the CDHs was lower among men than women. We conclude that the cumulative effects of indoor temperature exposure should be considered to reduce IRD risk in both cooling and heating seasons and NIRD risk in heating season and the cumulative effect on different gender.

Residential green space structures are associated with a lower risk of bipolar disorder: A nationwide population-based study in Taiwan.

Although many researchers have identified the potential psychological benefits offered by greenness, the association between green space structures and mental disorders is not well understood. The purpose of this study was to identify associations between green space structures and the incidence of bipolar disorder. To this end, we investigated 1,907,776 individuals collected from Taiwan's National Health Insurance Research Database. After a follow-up investigation from 2005 to 2016, among those with no history of bipolar disorder, 20,548 individuals were further found to be diagnosed with bipolar disorder. A geographic information system and landscape index were used to quantify three indices of green space structures: mean patch area (area and edge), mean fractal dimension index (shape), and mean proximity index (proximity). Additionally, greenness indices, the normalized difference vegetation index, and the enhanced vegetation index were used to confirm the association between greenness and incidence of bipolar disorder. These five indices were used to represent the individual's exposure according to the township of the hospital that they most frequently visited with symptoms of the common cold. Spearman's correlation analysis was performed to select variables by considering their collinearity. Subsequently, the frailty model for each index was used to examine the specific associations between those respective indices and the incidence of bipolar disorder by adjusting for related risk factors, such as socioeconomic status, metabolic syndrome, and air pollution. A negative association was identified between the mean patch area and the mean proximity index, and the incidence of bipolar disorder. In contrast, a positive association was found between the mean fractal dimension index and the incidence of bipolar disorder. We observed similar results in sensitivity testing and subgroup analysis. Exposure to green spaces with a larger area, greater proximity, lower complexity, and greener area may reduce the risk of bipolar disorder.

Associations among phthalate exposure, DNA methylation of TSLP, and childhood allergy.

BACKGROUND: Dysregulation of thymic stromal lymphopoietin (TSLP) expressions is linked to asthma and allergic disease. Exposure to phthalate esters, a widely used plasticizer, is associated with respiratory and allergic morbidity. Dibutyl phthalate (DBP) causes TSLP upregulation in the skin. In addition, phthalate exposure is associated with changes in environmentally induced DNA methylation, which might cause phenotypic heterogeneity. This study examined the DNA methylation of the TSLP gene to determine the potential mechanism between phthalate exposure and allergic diseases. RESULTS: Among all evaluated, only benzyl butyl phthalate (BBzP) in the settled dusts were negatively correlated with the methylation levels of TSLP and positively associated with children's respiratory symptoms. The results revealed that every unit increase in BBzP concentration in the settled dust was associated with a 1.75% decrease in the methylation level on upstream 775 bp from the transcription start site (TSS) of TSLP (ß = - 1.75, p = 0.015) after adjustment for child's sex, age, BMI, parents' smoking status, allergic history, and education levels, PM2.5, formaldehyde, temperature; and relative humidity. Moreover, every percentage increase in the methylation level was associated with a 20% decrease in the risk of morning respiratory symptoms in the children (OR 0.80, 95% CI 0.65-0.99). CONCLUSIONS: Exposure to BBzP in settled dust might increase children's respiratory symptoms in the morning through decreasing TSLP methylation. Therefore, the exposure to BBzP should be reduced especially for the children already having allergic diseases.

Spatial patterns of lower respiratory tract infections and their association with fine particulate matter.

This study aimed to identify the spatial patterns of lower respiratory tract infections (LRIs) and their association with fine particulate matter (PM2.5). The disability-adjusted life year (DALY) database was used to represent the burden each country experiences as a result of LRIs. PM2.5 data obtained from the Atmosphere Composition Analysis Group was assessed as the source for main exposure. Global Moran's I and Getis-Ord Gi* were applied to identify the spatial patterns and for hotspots analysis of LRIs. A generalized linear mixed model was coupled with a sensitivity test after controlling for covariates to estimate the association between LRIs and PM2.5. Subgroup analyses were performed to determine whether LRIs and PM2.5 are correlated for various ages and geographic regions. A significant spatial auto-correlated pattern was identified for global LRIs with Moran's Index 0.79, and the hotspots of LRIs were clustered in 35 African and 4 Eastern Mediterranean countries. A consistent significant positive association between LRIs and PM2.5 with a coefficient of 0.21 (95% CI 0.06-0.36) was identified. Furthermore, subgroup analysis revealed a significant effect of PM2.5 on LRI for children (0-14 years) and the elderly (≥ 70 years), and this effect was confirmed to be significant in all regions except for those comprised of Eastern Mediterranean countries.

Using a land use regression model with machine learning to estimate ground level PM2.5.

Ambient fine particulate matter (PM2.5) has been ranked as the sixth leading risk factor globally for death and disability. Modelling methods based on having access to a limited number of monitor stations are required for capturing PM2.5 spatial and temporal continuous variations with a sufficient resolution. This study utilized a land use regression (LUR) model with machine learning to assess the spatial-temporal variability of PM2.5. Daily average PM2.5 data was collected from 73 fixed air quality monitoring stations that belonged to the Taiwan EPA on the main island of Taiwan. Nearly 280,000 observations from 2006 to 2016 were used for the analysis. Several datasets were collected to determine spatial predictor variables, including the EPA environmental resources dataset, a meteorological dataset, a land-use inventory, a landmark dataset, a digital road network map, a digital terrain model, MODIS Normalized Difference Vegetation Index (NDVI) database, and a power plant distribution dataset. First, conventional LUR and Hybrid Kriging-LUR were utilized to identify the important predictor variables. Then, deep neural network, random forest, and XGBoost algorithms were used to fit the prediction model based on the variables selected by the LUR models. Data splitting, 10-fold cross validation, external data verification, and seasonal-based and county-based validation methods were used to verify the robustness of the developed models. The results demonstrated that the proposed conventional LUR and Hybrid Kriging-LUR models captured 58% and 89% of PM2.5 variations, respectively. When XGBoost algorithm was incorporated, the explanatory power of the models increased to 73% and 94%, respectively. The Hybrid Kriging-LUR with XGBoost algorithm outperformed the other integrated methods. This study demonstrates the value of combining Hybrid Kriging-LUR model and an XGBoost algorithm for estimating the spatial-temporal variability of PM2.5 exposures.

Ambient viral and bacterial distribution during long-range transport in Northern Taiwan.

Long-range transport (LRT) reportedly carries air pollutants and microorganisms to downwind areas. LRT can be of various types, such as dust storm (DS) and frontal pollution (FP); however, studies comparing their effects on bioaerosols are lacking. This study evaluated the effect of LRT on viral and bacterial concentrations in Northern Taiwan. When LRT occurred and possibly affected Taiwan from August 2013 to April 2014, air samples (before, during, and after LRT) were collected in Cape Fugui (CF, Taiwan's northernmost point) and National Taiwan University (NTU). Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) was applied to quantify influenza A virus. qPCR and qPCR coupled with propidium monoazide were, respectively, used to quantify total and viable bacteria. Types and occurrence of LRT were confirmed according to the changing patterns of meteorological factors and air pollution, air mass sources (HYSPLIT model), and satellite images. Two Asian DS and three FP cases were included in this study. Influenza A virus was detected only on days before and during FP occurred on January 3-5, 2014, with concentrations of 0.87 and 10.19 copies/m3, respectively. For bacteria, the increase in concentrations of total and viable cells during Asian DSs (17-19 and 25-29 November 2013) was found at CF only (from 3.13 to 3.40 and from 2.62 to 2.85 log copies/m3, respectively). However, bacterial levels at NTU and CF both increased during FP and lasted for 2 days after FP. In conclusion, LRT increased the levels of influenza A virus and bacteria in the ambient air of Northern Taiwan, particularly at CF. During and 2 days (at least) after LRT, people should avoid outdoor activities, especially in case of FP.

Chemical and stable isotopic characteristics of PM2.5 emitted from Chinese cooking.

This study investigated the characteristics of air pollutants generated from preparing Chinese cuisine and analyzed the isotopic compositions of carbon and nitrogen in particulate matter with a diameter <2.5 µm (PM2.5) to source apportionment study. The CO and CO2 concentrations and temperatures were measured using suitable instruments in real time during cooking, including stir-fry, fry, deep-fry, hot-pot, and mixed cooking, and periods with non-cooking. Personal environmental monitoring instruments were used to collect PM2.5 for carbon and nitrogen elements and isotopes analysis. Our data indicated that the concentrations of CO and CO2 and the temperature were higher during periods of cooking, especially for the fry and stir-fry methods, than during periods with non-cooking. The concentrations of PM2.5, total carbon, and total nitrogen were also higher during cooking; the maximum concentrations were measured during fry. The values of δ13C were considerably lower during the periods of cooking (mean: -28.15‰) than during non-cooking (-27.18‰). The average values of δ15N were 8.63‰ and 11.74‰ during deep-fry and hot-pot cooking, respectively. The δ13C values can be used to distinguish between cooking and other non-cooking sources and further assess the effect of different cooking activities on PM2.5. The δ15N only can be used to investigate the effect of deep-fry on PM2.5. Moreover, the δ13C signature suggested that fry emits higher products of incomplete combustion than do other cooking activities. These findings can assist in pollution source identification of PM2.5, emission control, and the study of combustion characteristics.

Comparison of Spatial Modelling Approaches on PM10 and NO2 Concentration Variations: A Case Study in Surabaya City, Indonesia.

Because of fast-paced industrialization, urbanization, and population growth in Indonesia, there are serious health issues in the country resulting from air pollution. This study uses geospatial modelling technologies, namely land-use regression (LUR), geographically weighted regression (GWR), and geographic and temporal weighted regression (GTWR) models, to assess variations in particulate matter (PM10) and nitrogen dioxide (NO2) concentrations in Surabaya City, Indonesia. This is the first study to implement spatiotemporal variability of air pollution concentrations in Surabaya City, Indonesia. To develop the prediction models, air pollution data collected from seven monitoring stations from 2010 to 2018 were used as dependent variables, while land-use/land cover allocations within a 250 m to 5000 m circular buffer range surrounding the monitoring stations were collected as independent variables. A supervised stepwise variable selection procedure was applied to identify the important predictor variables for developing the LUR, GWR, and GTWR models. The developed models of LUR, GWR, and GTWR accounted for 49%, 50%, and 51% of PM10 variations and 46%, 47%, and 48% of NO2 variations, respectively. The GTWR model performed better (R2 = 0.51 for PM10 and 0.48 for NO2) than the other two models (R2 = 0.49-0.50 for PM10 and 0.46-0.47 for NO2), LUR and GWR. In the PM10 model four predictor variables, public facility, industry and warehousing, paddy field, and normalized difference vegetation index (NDVI), were selected during the variable selection procedure. Meanwhile, paddy field, residential area, rainfall, and temperature played important roles in explaining NO2 variations. Because of biomass burning issues in South Asia, the paddy field, which has a positive correlation with PM10 and NO2, was selected as a predictor. By using long-term monitoring data to establish prediction models, this model may better depict PM10 and NO2 concentration variations within areas across Asia.

Development of Hourly Indoor PM2.5 Concentration Prediction Model: The Role of Outdoor Air, Ventilation, Building Characteristic, and Human Activity.

Exposure to indoor particulate matter less than 2.5 µm in diameter (PM2.5) is a critical health risk factor. Therefore, measuring indoor PM2.5 concentrations is important for assessing their health risks and further investigating the sources and influential factors. However, installing monitoring instruments to collect indoor PM2.5 data is difficult and expensive. Therefore, several indoor PM2.5 concentration prediction models have been developed. However, these prediction models only assess the daily average PM2.5 concentrations in cold or temperate regions. The factors that influence PM2.5 concentration differ according to climatic conditions. In this study, we developed a prediction model for hourly indoor PM2.5 concentrations in Taiwan (tropical and subtropical region) by using a multiple linear regression model and investigated the impact factor. The sample comprised 93 study cases (1979 measurements) and 25 potential predictor variables. Cross-validation was performed to assess performance. The prediction model explained 74% of the variation, and outdoor PM2.5 concentrations, the difference between indoor and outdoor CO2 levels, building type, building floor level, bed sheet cleaning, bed sheet replacement, and mosquito coil burning were included in the prediction model. Cross-validation explained 75% of variation on average. The results also confirm that the prediction model can be used to estimate indoor PM2.5 concentrations across seasons and areas. In summary, we developed a prediction model of hourly indoor PM2.5 concentrations and suggested that outdoor PM2.5 concentrations, ventilation, building characteristics, and human activities should be considered. Moreover, it is important to consider outdoor air quality while occupants open or close windows or doors for regulating ventilation rate and human activities changing also can reduce indoor PM2.5 concentrations.

Association between Surrounding Greenness and Mortality: An Ecological Study in Taiwan.

Exposure to surrounding greenness is associated with reduced mortality in Caucasian populations. Little is known however about the relationship between green vegetation and the risk of death in Asian populations. Therefore, we opted to evaluate the association of greenness with mortality in Taiwan. Death information was retrieved from the Taiwan Death Certificate database between 2006 to 2014 (3287 days). Exposure to green vegetation was based on the normalized difference vegetation index (NDVI) collected by the Moderate Resolution Imagine Spectroradiometer (MODIS). A generalized additive mixed model was utilized to assess the association between NDVI exposure and mortality. A total of 1,173,773 deaths were identified from 2006 to 2014. We found one unit increment on NDVI was associated with a reduced mortality due to all-cause (risk ratio [RR] = 0.901; 95% confidence interval = 0.862-0.941), cardiovascular diseases (RR = 0.892; 95% CI = 0.817-0.975), respiratory diseases (RR = 0.721; 95% CI = 0.632-0.824), and lung cancer (RR = 0.871; 95% CI = 0.735-1.032). Using the green land cover as the alternative green index showed the protective relationship on all-cause mortality. Exposure to surrounding greenness was negatively associated with mortality in Taiwan. Further research is needed to uncover the underlying mechanism.

Cumulative effect of indoor temperature on cardiovascular disease-related emergency department visits among older adults in Taiwan.

Studies have demonstrated that exposure to extreme outdoor temperatures increases cardiovascular disease mortality and morbidity. However, people spend 80%-90% of their time indoors, and the cumulative effects of exposure to high or low temperature on the risk of cardiovascular diseases had not been considered. This study investigated the cumulative effects of high or low indoor temperature exposure on the risk of cardiovascular diseases. We estimated indoor temperatures by using a prediction model of indoor temperature from a previous study and further calculated the cumulative degree hours at different indoor temperature ranges. Samples of emergency department visits due to cardiovascular diseases were collected from the Longitudinal Health Insurance Database (LHID) from 2000 to 2014 in Taiwan. We used a distributed lag nonlinear model to analyze the data. Our data demonstrated a significant risk of emergency department visits due to cardiovascular diseases at 27, 28, 29, 30, and 31 °C when cooling cumulative degree hours exceeded 62, 43, 16, 1, and 1 during the hot season (May to October), respectively, and at 19, 20, 21, 22, and 23 °C when heating cumulative degree hours exceeded 1, 1, 1, 11, and 33 during the cold season (November to April), respectively. Cumulative degree hours were different according to gender and age groups. Policymakers should further consider the cumulative effects to prevent hot- or cold-related cardiovascular diseases for populations.