SCH: INT: Individualizing Care in Pregnancy and Childbirth through Digital Phenotyping

点击翻译按钮获取中文摘要

Bringing together researchers and physicians from the Department of Women's Health, the Texas Advanced Computing Center, and the Institute for Computational Engineering & Sciences at The University of Texas at Austin, the goal of this project is to develop a digital phenotype of pregnancy to better understand factors influencing pregnancy outcomes. Women's health represents one of the most pressing health-policy issues impacting our nation. In no medical specialty are the deficiencies of medical evidence more pronounced than in women's health, especially in obstetrics. Over the course of the human life span, birth is one of the most dangerous health episodes for both mother and baby. Worldwide, between 2.6 and 4 million pregnancies result in stillbirth annually. Unlike other leading causes of mortality, birth-related deaths are largely preventable. Today, however, most adverse pregnancy outcomes are not predictable, and cannot be prevented. In this project, the research team will passively monitor a cohort of one thousand pregnant women from their first prenatal visit to six weeks post-partum. To accomplish this, participants will download the HealthyPregnancy smartphone app developed in this project to collect in situ social and behavioral data. The application passively captures participant's interactions with people and places via sensors and software throughout pregnancy. Analysis of this large collection of digital data, in combination with traditional medical monitoring data collected via participant's medical records will result in a digital phenotype of pregnancy. The digital phenotype allows for a more complete understanding of pregnancy at the macro scale and for more detailed understanding of outcomes as a continuum rather than isolated discrete events. It is widely understood that activity, social support, sleep, and cognitive function are important markers of health, particularly during pregnancy. Maternal obesity is associated with a number of complications in pregnancy including gestational diabetes, pre-eclampsia, macrosomia, caesarean delivery and stillbirth. Lack of social support and social interaction is also an important risk factor and has been shown to have adverse effects on pregnancy outcomes. Sleep disturbances are associated with poor health outcomes, particularly cardiovascular disease and inflammatory responses. Additionally, short sleep duration is associated with an increased incidence in diabetes and obesity and has been associated with an increase in mortality. Research suggests that women who experience pre-eclampsia more frequently report daily cognitive failures and increased emotional dysfunction years later. With the ubiquitous use of smartphones, it is now possible to collect lived experiences or data reflecting markers of pregnancy in the wild. Collecting accelerometer and GPS over time provide an indication of physical mobility and gross motor activity. Call and text message logs detail communication events and contribute to a view of social interaction and social contacts. Additionally, power state, screen time and touch events can be used to understand potential sleep disruption. Of greater significance is the analysis of this data in aggregate over the course of pregnancy. This longitudinal view and analysis of pregnancy in the wild using machine learning and mathematical models provides both an individual's digital phenotype of pregnancy and an aggregate digital phenotype of pregnancy. By gathering and analyzing these two products, they can be used to better understand outcomes and the continuum of events leading up to pregnancy outcomes.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.