Advances in artificial intelligence (AI) are bringing incredible opportunities to almost every aspect of life. Carolina is already doing outstanding work with this burgeoning technology, including our deep well of expertise within the Department of Computer Science and at the Renaissance Computing Institute (RENCI). Along with the growth of the UNC School of Data Science and Society (SDSS), our expertise and capabilities continue to foster more collaboration than ever before. As AI begins to transform society, we are working to ensure Carolina researchers lead in the use of AI and digital technologies.
Indeed, we have a wealth of research on our campus that leverages AI – too much to fit into a single blog! Thus, I will focus on our work in the health and clinical domains where collaborations and technological prowess in AI combine with expertise in health to make significant advancements in health science research and applied clinical care.
With recent funding from the Bill & Melinda Gates Foundation, an interdisciplinary team of Carolina researchers is working to expand prenatal care access in low-resource settings like Zambia and rural North Carolina through low-cost, AI-assisted ultrasound devices. Jeffery Stringer, professor and division director of global women’s health in the UNC School of Medicine (SOM), and his team have created an AI model that accurately estimates fetal age from scans taken by untrained users on simple devices. This research will be foundational for developing comprehensive AI tools that will improve access and pregnancy care worldwide.
W.R. Kenan, Jr. Distinguished Professor of Biostatistics at the Gillings School of Global Public Health, Michael Kosorok works at the forefront of sophisticated methods in precision health and AI with applications designed to improve health care. His Precision Health and AI Research Lab develops new algorithms and statistical methodology for data-driven decision making and application across a wide variety of human health domains, assessing patient characteristics and treatment regimens rapidly. Kosorok has used his advanced precision health methods in collaboration with others across campus to address critical health issues like type I diabetes, cancer, osteoarthritis, and more.
K.H. Lee Distinguished Professor Alex Tropsha leads the Molecular Modeling Lab at the Eshelman School of Pharmacy. His lab works at the intersection of pharmaceutical development and data science to accelerate the discovery of novel drugs. Leveraging the power of AI, the team tests solutions for therapeutics for illnesses like Alzheimer’s, cancer, and viral diseases. He and his team also collaborate with the chemistry and applied physical sciences departments to support AI-accelerated discovery of solar energy materials
UNC School of Nursing Associate Professor Jessica Zegre-Hemsey uses data science methods to improve diagnostic accuracy in emergency cardiac diagnoses. Her work has shown that AI can determine the likelihood of time-sensitive conditions like acute coronary syndrome, resulting in quicker response and care in life-threatening situations.
The Biomedical Image Analysis Group in the Department of Computer Science within the College of Arts and Sciences focuses on the design of computational algorithms to extract quantitative measures from biomedical data, creating easier to understand diagnostics. Led by Professor Marc Niethammer, the group focuses on methods for statistical shape analysis, image segmentation, deformable image registration, and machine learning. Niethammer also provides faculty support to Carolina’s AI Project which addresses the utilization of AI and augmented/virtual reality through a philosophic lens.
Led by Corbin Jones, professor of biology within the College and genetics in SOM, the Creativity Hubs funded AIxB team is combining an AI approach with biological sciences to examine how genetic information becomes transcribed and is regulated at a chemical level – cracking codes of how DNA controls gene expression and ultimately cell functions without changing DNA sequence. This research can interpret the complexity of genetic codes and offer insight into regulatory mechanisms that lead to health and disease. This work has potential to reduce experimental time and eventually inform therapies for treatable DNA-based diseases.
There is also much expertise in SOM’s computational medicine department. For example, Oliver Smithies Investigator and Chair of Genetics Fernando Pardo-Manuel de Villena, leverages these tools to explore genetic diversity across several models to answer fundamental questions about genomics, systems genetics, and evolution to understand critical biological processes to improve health.
Last year, SDSS gained its first cohort of faculty members. Many of these new faces have joint appointments in departments outside of the school. Associate Professor Hsun-Ta Hsu, who you will hear more about soon in this month’s issue of Endeavors, is jointly appointed in the School of Social Work and analyzes the implication of using AI to address societal health issues, like homelessness.
RENCI Director Ashok Krishnamurthy, along with Karamarie Fecho and Sarah Tyndell, organized the inaugural Clinical and Environmental Health Data Workshop last May, which covered issues related to clinical informatics, biomedical data ecosystems, and cyberinfrastructure. The conference addressed silos that prevent organizations like the U.S. Environmental Protection Agency and National Institute of Environmental Health Sciences from integrating data that is crucial to improving health outcomes. This integrated data empowers researchers and officials to pinpoint environmental inequalities, delve into social health drivers, and tailor interventions for specific communities. This research equips policymakers to target aid, optimize resources, and ensure fairness in health care access.
Yet with all the promise of AI, there remain questions about its use (and potential misuse). It is critical that we use AI responsibly to assist discoveries and leverage it appropriately. Thankfully, Stan Ahalt, dean of SDSS, is leading the UNC Generative AI Committee and providing guidelines that help researchers navigate the challenges and opportunities of AI resources.
This blog only begins to scratch the surface. With literally dozens of experts in departments ranging from obstetrics to linguistics, the potential for AI to supercharge our research enterprise permeates throughout our campus. Even more opportunities are possible as we think about incorporating AI into research and education. I am excited for the collaborations and groundbreaking solutions that will yield as we consciously and efficiently integrate this tool into our efforts to improve lives and well-being.