Machine Learning Model Predicts Indoor Ozone Exposure with Unprecedented Accuracy

Understanding indoor ozone behavior is essential for accurate health risk assessment since people spend most of their time indoors. A new study published in Eco-Environment & Health on July 9, 2025, presents the first large-scale machine learning model capable of predicting hourly indoor ozone concentrations using easily accessible predictors. The research addresses a significant limitation in current exposure assessments, which typically rely on outdoor data despite people spending 70%–90% of their time indoors where ventilation, indoor sources, and building materials affect actual ozone levels.

Researchers from Fudan University and the Chinese Academy of Sciences developed random forest algorithms trained on over 8,200 hours of indoor ozone data collected from 23 households across 18 Chinese cities. The model incorporates outdoor ozone levels from high-resolution datasets, meteorological parameters including temperature, humidity, wind, solar radiation, boundary-layer height, and surface pressure, along with window-opening behavior recorded manually by volunteers. By comparing two models—one excluding and one including window status—the researchers demonstrated that incorporating ventilation behavior significantly improved prediction accuracy, raising cross-validation R² from 0.80 to 0.83 and lowering RMSE from 7.89 to 7.21 ppb.

The study, available at https://doi.org/10.1016/j.eehl.2025.100170, reveals that indoor ozone concentrations were 40% lower than outdoor levels during the day, highlighting the buffering effect of indoor environments. Predictor-importance analysis identified surface pressure, temperature, and ambient ozone as dominant factors, with ventilation emerging as a crucial behavioral determinant. The model performed better in southern than northern China and in cold rather than warm seasons, accurately capturing hourly ozone fluctuations and regional differences.

"Most exposure studies still rely on outdoor ozone data, but that's only half the story," said Prof. Xia Meng, senior author of the study. "Our findings show that ventilation behavior—something as simple as whether a window is open or closed—can change exposure dramatically. By integrating such behavioral data with meteorological information through machine learning, we can finally estimate indoor ozone more precisely at large scales." This advancement is particularly significant given that long-term ozone exposure contributed to nearly 490,000 deaths worldwide in 2021, according to global health data.

The research introduces a practical, low-cost strategy for predicting indoor ozone exposure in real time across large geographic areas. Traditional mechanistic models require detailed indoor parameters that are difficult to obtain in large-scale studies, while linear regression models struggle with nonlinear environmental relationships. This machine learning approach overcomes these limitations by using accessible environmental and behavioral data, making it scalable for widespread application. The model can be integrated into health-risk assessments, smart-home monitoring systems, and public-health surveillance platforms, enabling policymakers and scientists to better understand indoor-outdoor exposure differences.

Future applications could extend this framework to other pollutants such as fine particulate matter or nitrogen dioxide, incorporate smart sensors for automated window tracking, and expand monitoring to diverse climatic zones. This research represents a major step toward more realistic ozone exposure assessment, bridging environmental modeling with daily life and promoting healthier indoor environments in rapidly urbanizing regions. The work was funded by the National Natural Science Foundation of China, supporting the development of tools that strengthen epidemiological studies and help guide public-health interventions in urban and residential settings.