Our team is collaborating with physicians to develop mobile app platforms that use machine learning methods to derive valuable insights about human behavior and improve maternal health. This work will provide users with an individualized, interactive experience on their smartphone.
In the United States, 1 in every 10 infant births is premature, occuring earlier than 37 weeks of gestation. Newborns who are born prematurely are more likely to develop medical complications because their organs are under-developed and to spend time in the neonatal intensive care unit (NICU). In addition to being a stressful and emotionally taxing experience for the parents of premature babies, it is estimated that preterm birth costs $26B to the United States annually.
Key risk factors for preterm birth include: maternal weight gain, smoking, depression and poor attendance at prenatal appointments. Black women are 3-4 times more likely to have an early preterm birth compared to women of other ethnic groups. We designed an app to address various risk factors among vulnerable populations by providing women with feedback on their pregnancy and informing them about their health risks.
The pilot study, published in the Journal of Medical Internet Research mHealth and uHealth, designed to target preterm birth among a particularly difficult-to-reach population, was successful in extending obstetric care,. We are currently improving the precision of our app-based risk models.