Indra Neil Sarkar, Ph.D., M.L.I.S., director of biomedical informatics in the Vermont Center for Clinical and Translational Science at the University of Vermont (UVM) and Fletcher Allen Health Care, has been awarded a $35,000 international IBM Faculty Award in support of his efforts to collate data that eventually would allow clinicians to better treat mothers at high risk for preterm birth. The leading cause of death in newborns, preterm birth can cause serious health problems, developmental delays and disabilities such as cerebral palsy in babies who survive.
“Preterm birth stumps every investigator going into this space,” says Sarkar, an assistant professor of microbiology and molecular genetics at the UVM College of Medicine and assistant professor of computer science in UVM’s College of Engineering and Mathematical Sciences. “We don’t have a complete picture of what causes it, we have a hard time predicting it, and it’s a major health issue in this country, leading to high mortality and morbidity rates in newborns and costing at least $26 billion a year. It’s a huge problem for the U.S.”
The United States is unique among industrialized nations for its high preterm birth rate, at 11.5 percent. It ranks sixth in the world for preterm births, higher than many African countries, including the Democratic Republic of Congo, Ethiopia, Uganda and Sudan, according to the March of Dimes.
Sarkar hopes that his project leads to two outcomes: one, a set of clinical guidelines that help identify at-risk mothers early in their pregnancies so providers can better manage their care, and two, a comprehensive map of physical, genomic and environmental data that shed light on the diseases and conditions that might put a mother at risk for preterm birth.
“Ultimately, it could be a visual map, with tables and numbers,” he explained. “You could take a condition, such as diabetes, and understand what’s the impact of having diabetes on the risk for preterm birth, or having diabetes and being overweight, or having diabetes and drinking alcohol, or having all three.”
The idea grew out of Sarkar’s conversations with Elizabeth Bonney, M.D., professor of obstetrics, gynecology and reproductive sciences, about exploring ways to study preterm birth. This grant will allow him to build on his collaborations with Bonney, as well as Elizabeth Chen, Ph.D., assistant professor of medicine and associate director of biomedical informatics in the Center for Clinical and Translational Science.
Working with the Vermont Oxford Network and other institutions throughout New England, Sarkar will identify information that can be used in the study, drawing on national as well as regional and state data. He will also collaborate with members of the Healthcare Analytics Research group at the IBM T.J. Watson Research Center in Yorktown Heights, N.Y., to develop and test data mining algorithms for identifying potential correlations of interest from datasets that are gathered related to preterm birth.
“The data are generally available as an artifact of normal care,” he says. “Computers can be used to mine electronic health data. If designed correctly, a program can go in and identify a patient you want to study. You don’t look at every pregnant mom; you look at only those eligible, and you never need to see their name.”
Sarkar will design computer algorithms to sort through these millions of bits of so-called “big data” to determine “is it actually knowledge or is it just noise? That’s the huge challenge we’re up against.”
The grant will be used to buy computer equipment, software and data sets, and IBM will share its expertise and software tools.
The IBM Faculty Awards – presented annually – is a competitive worldwide program intended to foster collaboration between researchers at leading universities worldwide and those in IBM research, development and services organizations; and promote courseware and curriculum innovation to stimulate growth in disciplines and geographies that are strategic to IBM. Awardees must be nominated by an IBM employee (Sarkar was nominated by a colleague from the Healthcare Analytics Research group at the IBM T.J. Watson Research Center). To qualify for this internationally competitive award, the nominee must be a full-time professor at an accredited university which has a Ph.D. or M.B.A. program in the nominee’s field. Candidates must have an outstanding reputation for contributions in their field or, in the case of junior faculty, show unusual promise.