UMIACS Faculty Secure Funding for Projects Combining AI and Medicine

Apr 22, 2020

Three faculty members in the University of Maryland Institute for Advanced Computer Studies (UMIACS) have received funding to jumpstart research that combines powerful artificial intelligence tools with medical expertise in areas like aging, traumatic brain injury, mental illness, and more.

Michael Cummings, a professor of biology and director of the Center for Bioinformatics and Computational Biology, Philip Resnik, a professor of linguistics, and John Dickerson, an assistant professor of computer science, are involved with two of the four projects recently funded under a new initiative known as AI + Medicine for High Impact (AIM-HI).

The program—which joins faculty from the University of Maryland, College Park, with medical experts at the University of Maryland, Baltimore—provides up to $200K per year for each project, for a maximum of three years.

Officials hope that these cross-institutional efforts will lead to scientific breakthroughs with the potential to secure larger federal awards as the research progresses.

“There is tremendous promise in each of the awards to attract additional external funding while making an impact to the state and region,” says Eric Chapman, the associate vice president for research development at the University of Maryland, College Park. “By leveraging our complementary cross-campus expertise, the state of Maryland will be able to tackle major health issues of societal importance.”

Cummings will work with experts in pharmaceutical sciences and anesthesiology at UMB to study aging and traumatic brain injury in mice. The researchers will develop an analytical framework to identify predictive functional relationships between changes in different metabolic parameters during aging.

The ultimate goal, Cummings says, is to generate testable hypotheses about mechanisms contributing to aging under both normal and disease conditions, which could lead to identification of appropriate interventions.

The researchers hope to transfer their initial findings to similar conditions seen in humans, which might eventually contribute toward treatments for age-related diseases like dementia and Alzheimer’s.

Resnik and Dickerson are partnering with Carol Espy-Wilson, a professor of electrical and computer engineering at UMD, and psychiatric experts at the School of Medicine in Baltimore lead by professor Deanna Kelly. The group is developing machine learning algorithms that can help assess mental illness online.

The goal of the project, Resnik says, is to establish a novel technological framework that integrates speech, video and text analysis in order to identify, assess and prioritize at-risk individuals.

The team’s initial research will focus on schizophrenia, major depressive disorder, and suicidal behaviors, but ultimately, they aim to develop a framework that can be broadly applied in other areas of healthcare.

Story by Maria Herd