Integrating Artificial Intelligence into Causal Research in Epidemiology

Current Epidemiology Reports, Год журнала: 2025, Номер 12(1)

Опубликована: Март 24, 2025

Язык: Английский

---

Understanding the spatial and seasonal variation of the ground-level ozone in Southeast China with an interpretable machine learning and multi-source remote sensing

The Science of The Total Environment, Год журнала: 2024, Номер 917, С. 170570 - 170570

Опубликована: Янв. 29, 2024

---

Advanced Deep Learning Model for Predicting Water Pollutants Using Spectral Data and Augmentation Techniques: A Case Study of the Middle and Lower Yangtze River, China

Process Safety and Environmental Protection, Год журнала: 2025, Номер unknown, С. 107058 - 107058

Опубликована: Март 1, 2025

Язык: Английский

---

Combined PMF modelling and machine learning to identify sources and meteorological influencers of volatile organic compound pollution in an industrial city in eastern China

Atmospheric Environment, Год журнала: 2024, Номер 334, С. 120714 - 120714

Опубликована: Июль 23, 2024

Язык: Английский

---

Development of over 30-years of high spatiotemporal resolution air pollution models and surfaces for California

Environment International, Год журнала: 2024, Номер 193, С. 109100 - 109100

Опубликована: Окт. 28, 2024

Язык: Английский

---

Quantitative Analysis of Multi-Elements in a Micron-Sized Single Particle Based on Laser-Induced Breakdown Spectroscopy Signal Enhancement of an Optical Fiber Collimated System

Analytical Chemistry, Год журнала: 2025, Номер unknown

Опубликована: Янв. 15, 2025

With rapid, energy-intensive, and coal-fueled economic growth, global air quality is deteriorating, particulate matter pollution has emerged as one of the major public health problems worldwide. It extremely urgent to achieve carbon emission reduction prevention control, aiming at common problem weak unstable signals characteristic elements in application laser-induced breakdown spectroscopy (LIBS) technology for trace element detection. In this study, influence optical fiber collimation signal enhancement method on LIBS was explored. Then, system based an collimated spectral intensity signal-to-noise ratio (SNR) compared, influences different preprocessing methods variable selection prediction performance random forest (RF) calibration model were investigated. Finally, Savitzky–Golay convolution derivative (SG)-variable importance projection (VIP)-mutual information (MI)-RF (Zn), first-order (D1st)-variable measurement (VIM)-successive projections algorithm (SPA)-RF (Cu), D1st-VIM-MI-RF (Ni) optimal models constructed according hybrid method. The performances their RF after SG-VIP-MI D1st-VIM-SPA D1st-VIM-MI are presented follows: Zn (Rp2 = 0.9860; MREP 0.0590), Cu 0.9817; 0.0405), Ni 0.9856; 0.0875). above results demonstrate that method, strategy overcome key low SNR quantitative accuracy single particle expected provide a theoretical basis technical support situ online rapid monitoring matter.

Язык: Английский

---

Advancing Source Apportionment of Atmospheric Particles: Integrating Morphology, Size, and Chemistry Using Electron Microscopy Technology and Machine Learning

Environmental Science & Technology, Год журнала: 2025, Номер unknown

Опубликована: Фев. 4, 2025

To further reduce atmospheric particulate matter concentrations, there is a need for more precise identification of their sources. The SEM-EDS technology (scanning electron microscopy and energy-dispersive X-ray spectroscopy) can provide high-resolution imaging detailed compositional analysis particles with relatively stable physical chemical properties. This study introduces an advanced source apportionment pipeline (RX model) that uniquely combines computer-controlled scanning computer vision machine learning to trace particle sources by integrating single-particle morphology, size, information. In the evaluation using virtual data set known contributions, RX model demonstrated high accuracy, average errors 0.60% number 1.97% mass contribution. Compared balance model, model's accuracy stability improved 75.6 73.4%, respectively, proved effective in tracing Fe-containing atmosphere steel city China. indicates morphology serve as feature determining its source. findings highlight potential coupled techniques enhance our understanding pollution sources, offering valuable insights PM health risk assessment evidence-based policy-making.

Язык: Английский

---

Machine learning analysis of drivers of differences in PAH content between PM1 and PM10 in Zagreb, Croatia

Atmospheric Pollution Research, Год журнала: 2025, Номер unknown, С. 102541 - 102541

Опубликована: Апрель 1, 2025

Язык: Английский

---

Enhancing Differentiation of Oxygenated Organic Aerosol: A Machine Learning Approach to Distinguish Local and Transboundary Pollution

ACS ES&T Air, Год журнала: 2025, Номер 2(5), С. 891 - 902

Опубликована: Апрель 15, 2025

Accurate source apportionment of particulate matter (PM), especially organic aerosol (OA), is crucial for targeted mitigation efforts. Positive Matrix Factorization (PMF) powerful in attribution primary OA (POA); however, it often struggles to differentiate sources oxygenated (OOA) due their similar chemical profiles. In this study, a support vector regression machine learning (ML) model was developed enhance the OOA Dublin from 2016 2023. Rolling PMF analysis identified four POA factors and differentiated into less- more-oxidized (LO-OOA MO-OOA), highlighting significant role (47-74% total OA). The ML further distinguished locally produced (LO-OOAlocal MO-OOAlocal) transboundary transport exhibited robust performance across different pollution scenarios. relative importance revealed that LO-OOAlocal more impacted by fossil fuel emissions like hydrocarbon-like (20%) coal (14%), whereas MO-OOAlocal most influenced LO-OOA (17%), providing insights formation mechanisms. During mixed episode, results show despite contribution transport, local heating were critical OA, with accounting 68% reaching 78% during hours. These findings highlight ongoing need reduce achieve cleaner air Dublin. model's ability quantitatively separate offers invaluable future quality regulations.

Язык: Английский

---

AI and Machine Learning for Optimizing Waste Management and Reducing Air Pollution

Journal of Hazardous Toxic and Radioactive Waste, Год журнала: 2025, Номер 29(3)

Опубликована: Апрель 21, 2025

Язык: Английский

---