As part of our “Meet Yale Internal Medicine” series, today’s feature is on Dennis Shung, MD, MHS, PhD, assistant professor of medicine (digestive diseases), and director of Digital Health, Digestive Diseases.
Dennis Shung, MD, MHS, PhD, is an assistant professor of medicine interested in applying machine learning to the treatment and prevention of gastrointestinal (GI) diseases. While passionate about working at the intersection of medicine and technology, Shung’s journey into this field has been a winding one, and he is grateful for the family members and mentors who have helped guide him to where he is today.
In 2022, the Department of Internal Medicine awarded Shung the Iva Dostanic Physician-Scientist Trainee Award to recognize his passion for science and his exceptional contributions to the GI field, including the machine learning model developed by Shung that helps clinicians all over the world identify very low-risk patients with GI bleeding. The manuscript for this study was published in Gastroenterology, the most prominent journal in the field.
As an undergraduate student at Rice University, Shung was interested in studying the humanities. “Then I went home after my first semester, and my parents told me, ‘We’re not paying for your Hispanics Studies degree, so you have to figure something else out’,” he said. In an act of “rebellion” he pursued and received dual bachelor’s degrees - a Bachelor’s of Arts in Hispanics Studies and a Bachelor’s of Science in Biochemistry and Cell Biology. Fortunately, he has found that medicine aligns seamlessly with the humanities, and especially how both disciplines connect you with different people and cultures.
Shung’s work ethic was largely influenced by his mother and father, who immigrated to the U.S. from Taiwan and Hong Kong, respectively. “They really instilled in me from an early age that working hard is part of the American dream,” he said. “My mom is an accountant, and my dad is an engineer; they’re very much a solid middle-class family. They told me, ‘You have to work for everything. Don’t take anything for granted.’”
Focus on Machine Learning
It wasn’t until Shung came to Yale School of Medicine for his Internal Medicine residency and GI fellowship that he became interested in machine learning. He was friends with several data science PhD students, and they told him that “this was the new electricity,” said Shung. “I became so convinced, I was willing to go back to school and start learning the area from the ground up.’”
Captivated by the idea of combining machine learning with medicine, Shung consulted with Digestive Diseases Section Chief Loren Laine, MD, professor of medicine (digestive diseases), to develop a plan to reach his career goals. Laine, a world-renowned expert on gastrointestinal bleeding and upper gastrointestinal tract injury, agreed that there was a lot of potential for the application of machine learning to GI bleeding. He told Shung that he needed to become an expert in that area by performing a comprehensive systematic review of the topic.
“Loren was very supportive as a research mentor, and incredibly helpful in ensuring that I understood the contours of a potential research career,” said Shung. “One of the things that I learned very early on was that you need a team of mentors. I had a lot of different people who helped me navigate my interests.”
Among Shung’s mentors was Allen Hsiao, MD, professor of pediatrics (emergency medicine) and chief medical informatics officer of Yale New Haven Health, who helped Shung think through the process of incorporating informatics into his research career, encouraging him to approach research from a technical perspective in addition to a clinical one. Shung recalled that Fred Gorelick, MD, Henry J. and Joan W. Binder Professor of Medicine (digestive diseases) and of Cell Biology, was the first person to suggest that he pursue a PhD to develop a more structured way of thinking and to deepen his skils. Finally, James Boyer, MD, Ensign Professor of Medicine (digestive diseases), offered a crucial piece of advice: have one fundamental question that unifies all of your work.
“All of these different mentors at Yale really helped me chart a path forward,” said Shung. “These conversations helped me understand that in order to get where I wanted to be, I needed to get a master of health science in clinical informatics, then progress to a PhD with a focus on machine learning techniques. I found that I needed to be a content expert as well as a methodologist who designs validation studies and uses novel tools to apply them to clinically impactful problems.”
Shung earned an MHS in Clinical Informatics at Yale School of Medicine in 2020 and a PhD in Investigative Medicine from Yale Graduate School of Arts and Sciences in 2022 under the mentorship of Smita Krishnaswamy, PhD.
“I would say that my entire scientific career started during my master’s program,” he said. “I took courses with Yale undergrads, starting from statistics and probability to introduction to machine learning, to introduction to data science, and learned everything basically from scratch.” His time in the Krishnaswamy Lab exposed him to different tools that could be used to model risk, from diffusion maps to signal processing on knowledge graphs.
‘One Unifying Question’
Inspired by Boyer’s “one unifying question” philosophy, Shung has anchored his research around the question, “How can I keep low-risk patients out of the hospital?” He is motivated to develop and implement the most effective tools for predicting outcomes in GI bleeding, help providers make the best decisions for their patients, and improve the patient experience overall. “As I progressed through the PhD program and thought about the applicability of these tools across other disciplines and questions, I began to broaden my scope to include other challenges. But I started with that one question, and I’ve been continually asking it,” he said.
Shung reflects that he has come a long way since his first semester in college, and that the journey has been full of unexpected twists and turns. “This work is not something that I would have ever dreamed of even five years ago,” he said. “I would never have thought of doing a PhD in my life. So this has all been a surprise to me.” Currently, he is working with Jas Sekhon, PhD, Eugene Meyer Professor of Political Science and Statistics & Data Science in the Department of Statistics and Data Science at Yale. His work with Sekhon is on applying novel random forest algorithms to risk stratification and using causal inference tools to quantify treatment effects of different transfusion policies in patients with upper gastrointestinal bleeding.
Outside of work, Shung enjoys spending time with his wife and two children. “We attend Trinity Baptist church together, which has become a family away from family,” Shung said. “When we came from Texas eight years ago, New Haven was a new place for my wife and I, but over time, it has become our home and community.”
Since forming one of the nation’s first sections of hepatology and then gastroenterology over 50 years ago, Yale’s Section of Digestive Diseases has had an enduring impact on research and clinical care in gastrointestinal and liver disorders. To learn more about their work, visit Digestive Diseases.