ASPIRE 2026 AI-Driven Big Data Study Reveals Link Between PM2.5 Pollution and Declining ART Success Rates.

Collaborative research by Taiwan IVF Group and Stanford University experts identifies 5–12% drop in pregnancy rates due to air pollution.

A breakthrough artificial intelligence (AI)–assisted big data analysis study presented at the 2026 Congress of the Asia Pacific Initiative on Reproduction (ASPIRE) in Beijing identified significant associations between ambient PM2.5 air pollution exposure and reduced assisted reproductive technology (ART) success rates across the United States and Japan.

The research utilized AI-assisted integration and analysis of large environmental and reproductive medicine datasets using machine learning–supported biomedical analytics. Jeffrey Zi Kang Huang, a 16-year-old research intern and student at Taipei American School, lead the study and contributed to environmental dataset integration and AI-assisted biomedical data analysis as part of the research team.

The study was conducted by Taiwan IVF Group and Ton Yen General Hospital, Taiwan, in collaboration with Stanford University professors. The principal investigator of the project was Barry Behr, HCLD, Professor in the Department of Obstetrics and Gynecology at Stanford University.

Using ART outcome data from the United States and Japan between 2010 and 2022 integrated with WHO and NASA environmental datasets, the investigators identified consistent associations between higher PM2.5 exposure and lower ART success rates. Regions with the highest pollution exposure demonstrated approximately 5–12% lower clinical pregnancy rates and 4–10% lower live-birth rates compared with regions with lower pollution exposure.

Professor Barry Behr stated:

 “Artificial intelligence and large-scale biomedical data integration are becoming increasingly important tools in reproductive medicine research.”

Jeffrey Zi Kang Huang added:

 “This research experience showed me how artificial intelligence and big data analysis can be applied to real-world biomedical and environmental health problems. It was exciting to participate in integrating large international datasets and identifying patterns that may have important implications for reproductive medicine and public health.”

The investigators emphasized that future studies incorporating advanced AI predictive modeling and individual-level environmental exposure analysis may further clarify the relationship between air pollution and reproductive outcomes.