Learning public health systems in infectious disease management – How can data-driven optimization help?
Abstract: Agency for Healthcare Research and Quality (AHRQ) defines a learning health system as a health system in which internal data and experience (e.g., semi-mechanistic model) are systematically integrated with external evidence (e.g., observational data), and the leant knowledge is put into practice. Becoming a learning system is increasingly an imperative in healthcare delivery and public health. Among the many challenges faced by learning systems is the one challenge on effective use of data, i.e., integrating newly acquired data with existing mechanistic understanding of the system to update intelligence for better controlling the system. This becomes more prevalent as we come out from the COVID-19 pandemic. How can data-driven optimization help? In this talk, we consider the context of multi-period location-specific resource allocation in infectious disease management. We propose two algorithmic approach to solve multi-period decision problems requiring online training and re-optimization.
Bio: Dr. Nan Kong is Professor and Interim Head of Weldon School of Biomedical Engineering at Purdue. He was the former Associate Director for Health Systems at Purdue’s Regenstrief Center for Healthcare Engineering. He graduated with B.S. in Automation from Tsinghua University in 1999 and Ph.D. in Industrial Engineering from University of Pittsburgh in 2006. He joined Purdue BME in 2007. His primary research is innovating data-driven optimization and analytics to address challenges in health care delivery.
