188宝金博页面版

Machine learning-driven framework for realtime air quality assessment and predictive environmental health risk mappingM. Rajesh 1 , R. Ganesh Babu 2 , Usha Moorthy 3? & Sathishkumar Veerappampalayam Easwaramoorthy 4This research introduces a practical and innovative approach for real-time air quality assessment and health risk prediction, focusing on urban, industrial, suburban, rural, and traf f i c-heavy environments. The framework integrates data from multiple sources, including fi xed and mobile air quality sensors, meteorological inputs, satellite data, and localised demographic information. Using a combination of machine learning techniques such as Random Forest, Gradient Boosting, XGBoost, and Long Short-Term Memory (LSTM) networks the system predicts pollutant concentrations and classif i es air quality levels with high temporal accuracy. Interpretability is achieved through SHAP analysis, which provides insight into the most inf l uential environmental and demographic variables behind each prediction. A cloud-based architecture enables continuous data fl ow and live updates through a web dashboard and mobile alert system. Visual risk maps and health advisories are generated every fi ve minutes to support timely decision-making. The framework not only forecasts pollution trends but also identif i es vulnerable populations through spatial overlays. Future validation will include real-world sensor deployment and comparison with health impact records to ensure both scientif i c accuracy and community relevance.Keywords Real-time air quality forecasting, Environmental health risk mapping, Machine learning in environmental monitoring, Spatial-temporal pollution modeling, Demographic vulnerability assessmentAir quality degradation is increasingly recognised as a silent yet pervasive threat to both human health and environmental stability. Across many regions of the world especially densely populated urban centresair pollution has escalated beyond seasonal concern into a chronic public health issue. Numerous studies by environmental and health organizations have underscored the severe consequences of prolonged exposure to airborne pollutants, including respiratory illnesses, cardiovascular complications, and impaired cognitive development in children. Traditionally, air quality has been monitored using static ground-based stations that provide periodic readings of pollutants such as PM 2.5 , PM 10 , NO 2 , CO, and O 3 . While these stations deliver accurate results at localised points, they fall short of representing the broader spatial variability within a city or region. Th ese systems are limited due of high running expenses, limited dissemination, and delayed reporting. Public warnings and real-time danger prevention are hindered by the data availability delay. Technological advancements are being utilised by academics and policymakers in response to the increasing demand for accurate and fast data on air quality. Machine learning is a promising method because it can handle large and complex information and deliver valuable prediction insights. While machine learning shows great promise in many fi elds, it has yet to fi nd widespread application in environmental health risk mapping.T h ere are millions of people whose health is jeopardised by air pollution. Healthcare preventive, city planning, and early warning systems can all benef i t from health ef f ect prediction and real-time air quality assessment. Wherever there is a lack of early response, the exposure of youngsters, the elderly, and persons with pre-existing health conditions to dangerous amounts of pollution is a serious problem. Th is challenge is driving 1 Department of Computer Science and Engineering, Aarupadai Veedu Institute of Technology, Vinayaka Mission’s Research Foundation (DU), Paiyanur, Tamilnadu, India. 2 Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh 522 302, India. 3 School of Computer Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Bengaluru, Manipal, Karnataka, India. 4 School of Engineering and Technology, Sunway University, No. 5, Jalan Universiti, Bandar Sunway, 47500 Petaling Jaya, Selangor Darul Ehsan, Malaysia. ? email: m.usha@manipal.eduOPENScientif i c Reports | (2025) 15:28801 1 | https://doi.org/10.1038/s41598-025-14214-6www.nature.com/scientificreports