EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Hybrid Transfer Learning Approach for Ambient PM2.5 Exposure Modeling in LMIC Contexts: A Case Study from Lima, Peru

Qiang Pu iD, Qingyang Zhu, Elvis Medina, Alan Llacza, Bryan Vu, Jianzhao Bi, Kenan Li iD, Laura Nicolaou iD, William Checkley, Stella Hartinger, Kyle Steenland iD, Yang Liu iD

ACS ES&T Air · 11.09.2026 · 3 (9) · 2279-2289

DOI: 10.1021/acsestair.6c00125 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACS ES&T Air, online erschienen am 18.08.2026. Verfasst von Qiang Pu und 11 weiteren Autor:innen. Beteiligte Einrichtung: College for Public Health and Social Justice, Saint Louis University, St. Louis, Missouri 63103, United States. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 73 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: FIC NIH HHS, Fogarty International Center, National Institute on Aging (NIA), National Institute of Environmental Health Sciences.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Exposure to fine particulate matter (PM2.5) contributes to millions of premature deaths annually worldwide. While epidemiological studies increasingly rely on geospatial technologies, such as satellite remote sensing, low-cost sensors, and chemical transport models, to quantify air pollution exposure, sparse monitoring in many low- and middle-income countries often introduces sampling bias, potentially skewing exposure estimates. We developed a hybrid and transfer learning approach in Lima, Peru, that integrates multiple geospatial technologies and machine learning to mitigate sampling bias and improve PM2.5 predictions. Our approach incorporated ground PM2.5 observations from reference-grade and low-cost monitors, satellite-retrieved aerosol optical depth, simulations from Weather Research and Forecasting model coupled with Chemistry, and additional pollution-related variables. We accounted for the sampling bias using a geo-envrionmental similarity-based transfer learning, leveraging training data from data-rich regions (the United States) to inform the data-poor target region (Lima). We then built an instance-weighted random forest model to predict daily PM2.5 concentrations across Lima from 2010 to 2023. The model achieved a cross-validated R2 of 0.76 and reduced potential overestimation in mountainous regions (average PM2.5 reduced from 34.19 to 29.23 μg/m3). This approach provides a scalable solution to support robust epidemiological research and public health planning across data-scarce settings.

Schlagwörter

Stichwörter der Autor:innen

LMICsPM2.5exposure modelsampling biastransfer learning

Autor:innen und Institutionen

  • College for Public Health and Social Justice, Saint Louis University, St. Louis, Missouri 63103, United States.
  • Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, Atlanta, Georgia 30322, United States.
  • National Meteorological and Hydrological Service of Peru, Lima 15072, Peru.
  • Department of Environmental Health, Harvard TH Chan School of Public Health, Boston, Massachusetts 02115, United States.
  • Department of Environmental and Occupational Health Sciences, University of Washington, Seattle, Washington 98195, United States.
  • Division of Pulmonary and Critical Care, School of Medicine, Johns Hopkins University, Baltimore, Maryland 21205, United States; Department of Environmental Health and Engineering, Bloomberg School of Public Health and…

ORCID-Profile: Qiang Pu, Kenan Li, Laura Nicolaou, Kyle Steenland, Yang Liu

Förderung

  • FIC NIH HHS
  • Fogarty International Center, National Institute on Aging (NIA), National Institute of Environmental Health Sciences

Zitieren

APA

Pu, Q., Zhu, Q., Medina, E., Llacza, A., Vu, B., Bi, J., Li, K., Nicolaou, L., Checkley, W., Hartinger, S., Steenland, K., & Liu, Y. (2026). A Hybrid Transfer Learning Approach for Ambient PM2.5 Exposure Modeling in LMIC Contexts: A Case Study from Lima, Peru. ACS ES&T Air, 3(9), 2279-2289. https://doi.org/10.1021/acsestair.6c00125

Vancouver

Pu Q, Zhu Q, Medina E, Llacza A, Vu B, Bi J, et al. A Hybrid Transfer Learning Approach for Ambient PM2.5 Exposure Modeling in LMIC Contexts: A Case Study from Lima, Peru. ACS ES&T Air. 2026;3(9):2279-2289. doi:10.1021/acsestair.6c00125

BibTeX
@article{Pu2026,
  author = {Pu, Qiang and Zhu, Qingyang and Medina, Elvis and Llacza, Alan and Vu, Bryan and Bi, Jianzhao and Li, Kenan and Nicolaou, Laura and Checkley, William and Hartinger, Stella and Steenland, Kyle and Liu, Yang},
  title = {A Hybrid Transfer Learning Approach for Ambient PM2.5 Exposure Modeling in LMIC Contexts: A Case Study from Lima, Peru},
  journal = {ACS ES&T Air},
  year = {2026},
  volume = {3},
  number = {9},
  pages = {2279--2289},
  doi = {10.1021/acsestair.6c00125}
}
Kostenlos starten →