From left, moderator Jinjun Xiong, PhD, and panelists Ram Samudrala, PhD; Scott T. Doyle, PhD; Marianthi Markatou, PhD; and Eun-Hye Enki Yoo, PhD.
Published April 3, 2025
How are Buffalo Translational Consortium researchers navigating the use of artificial intelligence (AI) in research? How is AI advancing translational science? And, in the words of one University at Buffalo researcher, in what ways are “whole new avenues of discovery” being opened using AI?
These questions were among those discussed during presentations and a panel discussion at the University at Buffalo Clinical and Translational Science Institute (CTSI) Annual Forum on March 19 at the Clinical and Translational Research Center. Among the highlights were presentations and panel discussions on the topic “Advancing Translational Science With Artificial Intelligence.”
Moderating the panel was UB Institute for Artificial Intelligence and Data Science (IAD) Director Jinjun Xiong, PhD, SUNY Empire Innovation Professor, Department of Computer Science and Engineering, School of Engineering and Applied Sciences. Xiong referred to the panelists as “pioneers of interdisciplinary research and using AI.”
Following individual presentations, the four presenters — Ram Samudrala, PhD, professor in the Department of Biomedical Informatics and chief of its Division of Bioinformatics, Jacobs School of Medicine and Biomedical Sciences, co-director, CTSI Informatics Core; Marianthi Markatou, PhD, SUNY Distinguished Professor, associate chair of research and healthcare informatics, Department of Biostatistics, School of Public Health and Health Professions; Eun-Hye Enki Yoo, PhD, professor, Department of Geography, College of Arts and Sciences; and Scott T. Doyle, PhD, associate professor, Department of Pathology and Anatomical Sciences, Jacobs School — joined together as panelists for a series of questions.
Samudrala discussed the Computational Analysis of Novel Drug Opportunities (CANDO) platform, which speeds targeted drug discovery. He noted technical, philosophical, monetary, and social challenges of working with CANDO. “Handling all that, and having models upon models, those are challenges,” he said. “We have benchmarks that we can measure [against], and we can ultimately see, if you make a drug, if it is not effective. Even if it takes a little while, the truth will come out.”
Markatou’s presentation highlighted the concept of “patient-centered AI.” When asked about that term, she explained that there is a “focus on the data, but it is beyond that — it is not just that.” Markatou added that “the delivery of healthcare is given based on the wants and needs of patients and their caregivers.” And while data is of paramount concern, she believes “when we look at things we should be looking at the whole person.”
Yoo’s ambitious work involves geospatial health and the study of air pollutants using air quality monitors. She also attempts to “capture human mobility, which is particularly relevant for environmental exposure assessment,” adding that “AI helps us to better design the study and then analyze and effectively capture what we need to know.“ Yoo carries her own air quality sensor with her, explaining, “In order to ask somebody to do so, I actually do it myself.”
Doyle identified the use of federated learning as influencing his study design, which he described as “the way your cell phone will learn your speech patterns and then suggest things in your text message. We are trying to do the same but with biomedical images and see how effective these are compared to the traditional [methods].” He also noted the value of self-awareness when evaluating a study: “Would I be comfortable being a participant in my study?”
Pictured during the “State of the CTSI” address: Timothy F. Murphy, MD, SUNY Distinguished Professor of medicine and director, UB Clinical and Translational Science Institute.
Near the end of the panel discussion, Xiong asked the four panelists to share lessons learned from their use of AI, and advice for audience members.
Samudrala said curiosity and optimism are important, but stressed the need for researchers to be very careful. “From an academic point of view, I think being curious and trying to do something for the greater good and for humanity is something to keep in mind. That should be [on the] horizon.”
Echoing his presentation, Doyle stressed the value of reading books and papers critically. He also urged prospective users to “play around with an AI system” they are considering. “It is a fun thing to do, but it will also give you an insight into how these systems work.”
“Keep an open mind, but think very carefully,” cautioned Markatou. “If you want to use something, try to understand precisely what it is doing.” She added that it is vital to understand the components of a system as well the challenges.
Yoo urged critical thinking: “Do not trust what AI said. Always validate and question.” In addition, she recommends “double and triple checking” results. “AI will not just solve your problem,” she said. “You have to be an active user.”
The 2025 forum also featured welcoming remarks from CTSI Director Timothy F. Murphy, MD, SUNY Distinguished Professor of medicine; and Allison Brashear, MD, MBA, UB’s vice president for health sciences and dean of the Jacobs School; a “State of the CTSI” talk from Murphy; and presentations from the winner and finalists of 2024 Buffalo Translational Consortium (BTC) Clinical Research Achievement Awards.
The awardees were introduced by BTC Clinical Research Achievement Awards Committee Chair Anne B. Curtis, MD, SUNY Distinguished Professor, Department of Medicine, Jacobs School.
From left, 2024 BTC Clinical Research Achievement Awards Top Award recipient Andrew H. Talal, MD, MPH, professor of medicine; BTC Clinical Research Achievement Awards Committee Chair Anne B. Curtis, MD, SUNY Distinguished Professor of medicine; 2024 BTC Clinical Research Achievement Awards Finalist Stanley A. Schwartz, MD, PhD, UB Distinguished Professor of medicine and pediatrics; and 2024 BTC Clinical Research Achievement Awards Finalist Eunice S. Wang, MD, associate prrofessor of medicine.