Objective: This study aimed to predict and analyze air pollution-related mortality rates in Bangkok, Thailand, using a comprehensive neural network (NN) model. Objectives included analyzing temporal dynamics, evaluating model effectiveness, and identifying the influential factors. Material and Methods: Daily air quality and mortality data from 2016 to 2020 were used. We employed recurrent neural networks (RNNs), long short-term memory (LSTM), and gated recurrent units (GRU) models to capture the complex relationships between six key pollutants and mortality. The best model was selected based on the lowest R2 and mean difference from the Bland–Altman analysis. SHAP feature importance and subgroup analyses for age (0–5, 5–60, and 60+ years) and cause of death (respiratory, circulatory, and other) were conducted. Results: Analysis included 170,612 mortality cases over 1,828 days, with a median=36 daily premature deaths. An LSTM model with a 23-day time lag demonstrated the highest predictive accuracy for overall mortality. Subgroup analysis identified different optimal models; an RNN model was best for the “Older Adult” subgroup, and an LSTM model was best for the “Respiratory” subgroup. For feature importance, relative humidity, particulate Matter (PM2.5), and ozone (O3) were the most influential for overall mortality. The most influential factors for older adults were PM10, PM2.5, and carbon monoxide (CO); for respiratory, PM2.5, nitrogen dioxide (NO2), and PM10 were the most influential. Conclusion: This study demonstrates NN’s potential in predicting air pollution-related mortality rates in Bangkok. Findings highlight the importance of considering temporal dynamics, subgroup-specific characteristics, and the key environmental factors in model development. These data-driven insights can inform public health policies and facilitate targeted interventions to mitigate the health impacts of urban air pollution.
Kanakorn Horsiritham, Natthaya Bunplod, P. Sirinara et al.· Journal of Health Science an...· 0 citations
Communities residing near petrochemical complexes are chronically exposed to complex volatile organic compound (VOC) mixtures, but routine surveillance is often restricted to limited sentinel panels, which may lead to the underestimation of true health risks. This study assessed age-stratified probabilistic carcinogenic and noncarcinogenic inhalation risks near the Map Ta Phut Industrial Estate, Thailand, the world’s eighth-largest petrochemical hub.
During 2021–2023, six routinely monitored VOCs, namely, benzene, toluene, ethylbenzene, xylene, 1,3-butadiene, and vinyl chloride, were measured at two residential stations and one industrial fence-line station. An expanded 20-VOC panel was additionally analyzed at the fence-line station to capture broader industrial emissions. In this study, Thai-specific anthropometric data and the carcinogenic equivalence quotient framework were integrated into 10,000-iteration Monte Carlo simulations to estimate the excess lifetime cancer risk (ELCR) and hazard index (HI) for preschool children, school-aged children, adolescents, and adults. In addition, the annual number of VOC-attributable excess cancer cases in Rayong Province was also estimated.
The mean ELCRs were low at the community stations (4.10 × 10
− 7
to 3.08 × 10
− 6
) but increased markedly at the fence-line station (3.18 × 10
− 5
to 2.32 × 10
− 4
) and increased further under the expanded 20-VOC panel (5.00 × 10
− 5
to 3.61 × 10
− 4
). While the mean HI remained below 1.0 for the routine panel at all the stations, it exceeded 1.0 for all the age groups under the expanded 20-VOC panel (ranging from 2.04 to 4.83). Preschool children had the greatest noncarcinogenic burden, with a 95th -percentile HI of 10.8. Benzene and ethylbenzene were the main carcinogenic contributors at the two community stations, whereas vinyl chloride dominated both the carcinogenic and noncarcinogenic risks at the industrial fence line. The estimated number of adult VOC-attributable excess cancer cases increased from 2.79 in 2015 to 3.15 in 2023, accounting for 0.36% of the regional cancer burden.
Limited routine VOC monitoring may substantially underestimate petrochemical-related carcinogenic and noncarcinogenic risks. Age-stratified probabilistic assessment using an expanded VOC panel revealed substantial fence-line risks driven by specialty monomers, highlighting preschool children as a priority population for strengthened surveillance and site-specific risk management.
P. Sirinara, P. Chanpiwat, Natthaya Bunplod et al.· BMC Public Health· 0 citations