Study of How Air Pollution Affects Pregnancy and Risk of Early Birth Using Machine Learning

What this Trial Is About

Researchers are investigating the impact of air pollution on pregnancy, focusing on preterm birth and other adverse birth outcomes. This study brings together experts in obstetrics, child health, epidemiology, climate science, and artificial intelligence to develop a deep learning model. The model uses extensive patient data from electronic health records at University College London Hospital to predict the risk and timing of preterm birth, aiming to identify how air pollution exposure affects pregnancy outcomes. The study will collect data from about 18,000 pregnant women who delivered at University College London Hospitals from 2019 onwards. It links detailed clinical records with information on ambient pollution exposure based on postcode and delivery date. Using this data, the deep learning model will analyze the sequence of events in a patients medical history and pregnancy to predict preterm birth risk and determine when pollution exposure has the most significant effect. Participants data will be reviewed retrospectively, without direct intervention, using electronic health records. The models accuracy in predicting preterm birth and assessing pollution impact will be evaluated over 36 to 42 months. The study also plans to work with policy groups and health authorities to share findings and develop public health messages to help pregnant women reduce pollution exposure. The research spans several years, with ongoing analysis and potential future guidance development.

Air Pollution and Pregnancy