新闻中心NEWS

返回首页

讲座:Optimal Stopping for Medical Treatment with Predictive Information: Theory and Application in ICU Mechanical Ventilation

发布者:人力资源办公室    发布时间:2020-05-08

题 目:Optimal Stopping for Medical Treatment with Predictive Information: Theory and Application in ICU Mechanical Ventilation

嘉 宾:Zhichao Zheng, Associate Professor, the Singapore Management University

主持人:李成璋   助理教授   上海交通大学安泰经济与管理学院

时 间:2020年5月20日(周三)14:00-15:30

会议方式:ZOOM会议(校内师生如需会议号和密码,请于5月19日中午12点前发送电邮至mliu18@sjtu.edu.cn获取)

内容简介:

  Data availability and advancement in machine learning techniques make an accurate prediction of the future a foreseeable reality. How to efficiently incorporate the predictive information into a multistage medical decision-making environment, however, remains understudied. In this paper, we develop a discrete-time, finite-horizon Markov decision process (MDP) with predictions to support decisions on medical treatment continuation. We characterize the structure of the optimal policy and reveal important insights on how predictive information can lead to different decision protocols. We calibrate and apply our model to the mechanical ventilator extubation problem in an intensive care unit (ICU). Using a patient-level data set, we compare the performance of different policies and demonstrate that incorporating predictive information can reduce ICU length-of-stay (LOS) by up to 9.4% and meanwhile decrease the failure rate of ventilated patients by up to 18.9% even though the objective of the model is to minimize the expected LOS. The benefits are more significant for patients with poor initial conditions. Furthermore, simply optimizing LOS without using predictive information using a classical MDP model can lead to an increased failure rate of ventilated patients by up to 6%.  We also estimate using publicly available data on COVID-19 critical patients and show that applying the ventilation protocols using predictive information can increase the throughput of ICUs by up to 5%.

演讲人简介:

  Zhichao Zheng is an Associate Professor of Operations Management at the Singapore Management University. His main research interests lie in data analytics and optimization for healthcare operations management and medical decision making. He also applies his research in sharing economics and supply chain risk management, etc. He received his BS (First Class Honors) in Applied Mathematics from the National University of Singapore in 2009, and Ph.D. in Management from the Department of Decision Sciences (renamed to Department of Analytics & Operations) in the National University of Singapore in 2013.

欢迎广大师生参加!