Deep Representation Learning Identifies Associations Between Physical Activity And Sleep Patterns During Pregnancy And Prematurity

This study demonstrates the potential of using wearables data and computation to personalize and identify the likelihood of preterm birth. A dataset of 1083 pregnant individuals was collated and wearable actigraphy devices were given to them in their first trimester. Electronic health record data was collected along with physical activity data. The results suggest that consultation and intervention, informed by continuous monitoring, may enable cost-effective and scalable mitigations to the unacceptably high prevalence of preterm birth globally.