HAS Online Seminar

Health Applications Society Online Seminar Series

Welcome to
the HAS Online Seminar Series! This seminar series welcomes a broad range of healthcare modeling research topics such as healthcare operations, medical decision making, health policy, and health analytics. 

Stay Connected!

Please join our Google Group or subscribe to our Mailing List to receive the latest announcement on the upcoming seminars!
You can also add or download our seminar calendar (Google calendar or iCalendar)
If you have any questions, please feel free to reach us at hasseminarseries@gmail.com !

When: 1-2 pm ET (10-11 am PT) on 4th Friday of each month
Where: Zoom Webinar (Register Now!
View recording here

Speaker: Dr. Carri Chan, Columbia Business School

Seminar Title: Interpretable Machine Learning for Resource Allocation with Application to Ventilator Triage
Date and Time: July 22 (Friday), 2022, 1:00-2:00pm ET (10-11am PT)

: Rationing of healthcare resources is a challenging decision that policy makers and providers may be forced to make during a pandemic, natural disaster, or mass casualty event. Well-defined guidelines to triage scarce life-saving resources must be designed to promote transparency, trust and consistency. To facilitate buy-in and use during high stress situations, these guidelines need to be interpretable and operational.  We propose a novel data-driven model to compute interpretable triage guidelines based on policies for Markov Decision Process that can be represented as simple sequences of decision trees (tree policies).  In particular, we characterize the properties of optimal tree policies and present an algorithm based on dynamic programming recursions to compute good tree policies.  We utilize this methodology to obtain simple, novel triage guidelines for ventilator allocations for COVID-19 patients, based on real patient data from Montefiore hospitals. We also compare the performance of our guidelines to the official New York State guidelines that were developed in 2015 (well before the COVID-19 pandemic). Our empirical study shows that the number of excess deaths associated with ventilator shortages could be reduced significantly using our policy.  Our work highlights the limitations of the existing official triage guidelines, which need to be adapted specifically to COVID-19 before being successfully deployed. 

Carri W. Chan is the John A. Howard Professor of Business and the Faculty Director of the Healthcare and Pharmaceutical Management Program at Columbia Business School. Her research is in the area of healthcare operations management. Her primary focus is in data-driven modeling of complex stochastic systems, efficient algorithmic design for queuing systems, dynamic control of stochastic processing systems, and econometric analysis of healthcare systems. Her research combines empirical and stochastic modeling to develop evidence-based approaches to improve patient flow through hospitals. She has worked with clinicians and administrators in numerous hospital systems including Northern California Kaiser Permanente, New York Presbyterian, and Montefiore Medical Center. She is the recipient of a 2014 National Science Foundation (NSF) Faculty Early Career Development Program (CAREER) award, the 2016 Production and Operations Management Society (POMS) Wickham Skinner Early Career Award, and the 2019 MSOM Young Scholar Prize. She currently serves as a co-Department Editor for the Healthcare Management Department at Management Science. She received her BS in electrical engineering from MIT and MS and Ph.D. in electrical engineering from Stanford University.

Past Seminars

  • June 25, 2021: Dr. Alvin Roth, Stanford University, View Recording
      Abstract: Many patients in need of a kidney transplant have a willing but incompatible (or poorly matched) living donor. Kidney exchange programs arrange exchanges among such patient-donor pairs, in cycles and chains of exchange, so each patient receives a compatible kidney. Kidney exchange has become a standard form of transplantation in the United States and a few other countries, in large part because of continued attention to the operational details that arose as obstacles were overcome and new obstacles became relevant. We review some of the key operational issues in the design of successful kidney exchange programs. Kidney exchange has yet to reach its full potential, and the paper further describes some open questions that we hope will continue to attract attention from researchers interested in the operational aspects of dynamic exchange.
    • August 27, 2021: Dr. Edward Kaplan, Yale University, View Recording
        Abstract: This talk reviews some simple analyses that were developed in real time to support local decision-making. We start with Bernoulli models for capping the size of gatherings, and then consider applications of queueing models for assessing hospital ICU capacity. We then discuss using wastewater-based epidemiology to monitor local outbreaks. We also examine gateway and repeat viral testing to prevent coronavirus transmission on campus. Throughout we will illustrate how these analysis were used to inform local decisions.
      • September 24, 2021: Dr. Mark Van Oyen, University of Michigan, View Recording
          Abstract: Improving patient experience requires us to incorporate the perspective of the patient and their personalized care needs and desires. Examples include timely access to a future visit, personalized bed unit assignment, coordinated care, and personalized scheduling to care providers. Past models have usually emphasized system efficiency. It is important to include both efficiency and patient experience. Health systems are increasingly sensitive to patient experience, thereby presenting research opportunities. This talk emphasizes timely patient access/scheduling and either stratified or individualized care models for resource allocation. Recent advances are frequently too limited to allow direct application to healthcare for reasons such as personalization as well as the medical, ethical, organizational, and social dimensions. We present approaches to key problems using models with uncertainty/robustness, machine learning, online learning (e.g., multi-armed bandits), and their integration with optimization. Online methods offer a way to cope with limited historical data and to mitigate unpredictable shocks to the system, such as the COVID-19 pandemic.
        • January 28, 2022: Dr. John R. Birge, The University of Chicago Booth School of Business, View Recording
            Abstract: Many operations researchers have studied various issues concerned with controlling the COVID-19 pandemic (and, hopefully not too soon, future pandemics). This talk will discuss some of the areas in which OR has been applied such as forecasting and pharmaceutical and non-pharmaceutical mitigation measures. The discussion will consider the relative impact of these efforts and lessons for future research.
          • February 25, 2022: Dr. Ebru Bish, The University of Alabama Culverhouse College of Business, View Recording
              Abstract: The COVID-19 pandemic continues to demonstrate the importance of public health screening. My talk will draw upon the body of research that my collaborators and I have conducted in a variety of screening contexts, ranging from newborn screening for genetic disorders to population-level infectious disease screening, including COVID-19. I will discuss the challenges and opportunities for operations researchers.
            • March 25, 2022: Dr. Dávid Papp, North Carolina State University, View Recording
                Abstract: Optimization models and algorithms have played a critical role in radiotherapy treatments of cancer since the advent of intensity modulated radiotherapy in the 1990s, perfecting both the mathematical models and computational methods used in standard radiotherapy modalities. This talk will focus on another family of questions aimed at moving beyond conventional treatment planning: how optimization models can help with the selection, and possible combination, of treatment modalities, and in determining the ideal dose prescriptions for patients through a concept called biologically effective dose.
              • April 22, 2022: Dr. Andrew Li, Carnegie Mellon University, View Recording
                  Abstract: An accurate blood test for early-stage cancer (a “liquid biopsy”) is arguably the most important open problem in oncology, and the race to a solution is tantalizingly close to the finish. In this talk, we will discuss the state of this race as of 2022, particularly how technology and data have enabled progress so far, and how optimization will play a role in reaching the finish line.
                • May 27, 2022: Dr. Oguzhan Alagoz, University of Wisconsin-Madison, View Recording
                    Abstract: This talk describes the use of partially observable Markov decision processes (POMDPs) for personalizing cancer screening. POMDP models can be used to address several controversial open research questions in cancer screening, such as when to start and stop screening and how often to screen. We demonstrate the development and application of a POMDP-based personalized cancer screening policy using breast cancer as an example. In addition, we briefly describe how nonadherence to the screening recommendations, limited screening resources, and existence of chronic conditions could be addressed using the POMDP modeling framework. Finally, we describe successful POMDP applications in other cancers including personalizing colorectal and lung cancer screening.
                  • June 24, 2022: Dr. Vishal Gupta, USC Marshall School of Business, View Recording
                      Abstract: On July 1st, 2020, members of the European Union gradually lifted earlier COVID-19 restrictions on non-essential travel. In response, we designed and deployed “Eva” – a novel reinforcement learning system – across all Greek borders to identify asymptomatic travelers infected with SARS-CoV-2. Eva allocates Greece’s limited testing resources based on demographic characteristics and results from previously tested travelers to (i) limit the influx of new cases and (ii) provide real-time estimates of COVID-19 prevalence to inform border policies. Counterfactual analysis shows that Eva identified 1.85x as many asymptomatic, infected travelers as random surveillance testing, with up to 2-4x as many during peak travel. Moreover, Eva identified approximately 1.25-1.45x as many infected travelers as policies that require similar infrastructure as Eva, but make allocations based on population-level epidemiological metrics (cases/deaths/positivity rates) rather than reinforcement learning. This talk discusses some of the main design decisions behind Eva, the key elements of the reinforcement learning algorithm, and the measured impact of the system in the summer of 2020. 

                    Seminar Organizers and Advisory Board

                    This seminar series is organized by Sanjay Mehrotra (Northwestern University), Sait Tunc (Virginia Tech), Qiushi Chen (Pennsylvania State University).
                    The advisory board of the Year 2022 includes Mark Van Oyen (University of Michigan), Maria Mayorga (North Carolina State University), and Timothy Chan (University of Toronto).

                    Special thanks to INFORMS Health Applications Society and all board members for their enormous support!