Duke AI Health Welcomes Michael Zavlanos as Director in Healthcare System Optimization
“We are very excited about the unique insights Dr. Zavlanos will bring to challenges that directly affect the performance of our health system and the experiences of our patients and the clinicians and staff who care for them.”
– Duke AI Health Director Michael Pencina, PhD –
Duke AI Health has recently welcomed Michael Zavlanos, PhD, as Director in Healthcare System Optimization for Duke AI Health. In this newly created role, Dr. Zavlanos will direct efforts focused on developing and implementing machine learning, optimization, and other algorithmic tools to increase operational efficiency and resource use in health care at Duke.
Currently a professor in the Thomas Lord Department of Mechanical Engineering and Materials Science at Duke University’s Pratt School of Engineering and an Amazon Scholar with Amazon Robotics, Dr. Zavlanos received his doctorate in electrical and systems engineering from the University of Pennsylvania, where he specialized in robotics and autonomous control systems. More recently, at Duke, Dr. Zavlanos has begun to explore the potential for using data gathered from “real-world” clinical settings to solve problems with a direct impact for patients.
“We are very excited about the unique insights Dr. Zavlanos will bring to challenges that directly affect the performance of our health system and the experiences of our patients and the clinicians and staff who care for them,” said Michael Pencina, PhD, Chief Data Scientist for Duke Health and Vice Dean for Data Science at the Duke University School of Medicine.
Dr. Zavlanos’ role evolved from a series of collaborative, multidisciplinary projects focused on applying AI tools to solve complex challenges in optimizing scheduling and resource utilization for medical facilities and reducing patient wait times.
“One of the key problems faced by many hospitals and healthcare systems is how variable utilization – the demand for staffing and resources – is from day to day or week to week,” explains Zavlanos. “One of the projects we’ve been working on over the past couple of years suggests that the use of algorithmic technologies can help us better understand and control that variability, and in doing so increase the number of patients we can see each week while significantly reducing wait times for patients.”
Zavlanos notes that although efforts to improve health system efficiency using predictive models for case length or length of stay are not new, a systems-level approach that combines mathematical optimization and AI provides and powerful new tools to better understand and optimize the complex interactions among patients, providers, and resources that affect health system operations. Early efforts in these areas started with surgery scheduling for cardiothoracic surgery, neurosurgery, and ambulatory care, evaluated using retrospective data from electronic health records. Now, Dr. Zavlanos’ focus is widening to include the interplay between workflows and efficiency, as well as new collaborations across engineering, surgery, and anesthesiology.
“Things like scheduling and utilization may not seem like glamorous topics, but they are incredibly impactful at every level – patient wellbeing, working conditions for staff, operational capacity, finances, and more,” notes Duke University Health System Senior Vice President Richard Shannon, MD, who also serves as chief quality officer for Duke Health.
Jerome P. Lynch, PhD, Vinik Dean of Duke’s Pratt School of Engineering, notes that Zavlanos’ background makes him well suited for navigating the evolving challenges of the healthcare environment.
“The kinds of complex problems and dependencies that arise from scheduling and utilization in healthcare are ones that AI systems are well poised at helping us solve,” said Lynch. “This is where Dr. Zavlanos’ experience in thinking about ways to optimize performance in different systems, whether we’re talking about autonomous robotics or AI applications, will be invaluable to improving healthcare systems that lead to improved outcomes for patients.”
Dr. Zavlanos will continue to build connections between the Pratt School of Engineering and departments within the School of Medicine while focusing on research and methodological developments that are translatable to tangible improvements in health system operations.
Please join us in welcoming Dr. Zavlanos in his new role at Duke AI Health!