Cultivating a Community in Computational Medicine

For more than a decade, Amanda Randles, the Alfred Winborne and Victoria Stover Mordecai Associate Professor of Biomedical Sciences at Duke, has been developing supercomputer models of how blood flows through our arteries at the cellular and particulate levels.

Dubbed HARVEY for short, the evolving algorithm can simulate how an individual’s specific cardiovascular architecture will likely respond to surgical procedures such as a bypass graft or stent placement.

While impressive, this work is a small fraction of the research in the field of computational medicine that’s occurring at Duke. For example, Jessilyn Dunn, an assistant professor also in Duke BME, gathers biometric data from smartwatches and other wearable devices to detect patterns and develop models about a patient’s overall health.

Across the university, physicians in Duke’s School of Medicine develop robotic devices and virtual reality tools that enable surgeons to practice and plan complex surgeries.

While this work has been successful, it’s also been relatively siloed across departments and schools. Now, Randles and her colleagues are working together to transform Duke into a leader in the ever-advancing field.

“The field of computational medicine is very broad because the medical problems we can address with these tools and approaches are very broad,” explains Randles. “By working together and establishing a tight-knit community here at Duke, we can provide infrastructure and support for addressing major medical challenges that are more significant than one lab can take on.”

Over the summer of 2023, Randles organized a series of workshops for Duke faculty working within the world of computational medicine, with attendees joining from Duke’s Pratt School of Engineering, Trinity College of Arts and Sciences, School of Nursing, and from the divisions of Surgery, Cardiology and Anesthesiology within the Duke University School of Medicine.

The researchers had three goals: Develop grants to support research in computational medicine, identify and address the needs of the community, and identify the most pressing challenges in the field that the Duke community could work together to target and—hopefully—solve.

“We learned a lot about the capabilities of the different labs and the types of problems they were interested in solving, especially on the technology side. There is a big focus on advancing virtual reality, wearable technology and real-time computing,” says Randles. “But we also learned about the barriers they were facing, whether it was a computational problem or the need for additional training.”

Beyond giving the faculty an opportunity to address these goals, the workshops enabled the researchers to learn about other computational work that was occurring at Duke.

“We realized there was so much great work that was already being done, but it wasn’t being shared broadly across the university,” says Randles.

This realization led to the creation of the Computational Medicine Seminar Series. Officially launched in September 2023, the seminar series will feature a mix of speakers from Duke, who can share their research with the wider Duke community and potentially recruit interested students, and external experts, who can discuss their work to help illustrate the full breadth of the field. Maria Gorlatova, the Nortel Networks Assistant Professor of Electrical & Computer Engineering, served as the program’s inaugural speaker.

As the seminar series continues throughout the academic year, Randles and her collaborators will introduce additional community events, including coffee hours with faculty and students to discuss research, a research “speed dating” event to increase collaboration between Duke’s School of Medicine and Pratt School of Engineering, and skill-based training days focused on shared interest topics like wearable device development or using cloud computing.

All of these efforts will culminate in the launch of a new Duke research center dedicated to computational medicine.

“The depth of experience in computational medicine at Duke is incredibly impressive, and we want to make sure we’re doing everything to make sure our researchers are as successful as possible,” says Randles. “Our workshops and events so far have proven that there is a lot of enthusiasm and opportunity to make Duke a real leader in this area.”