Course Description

Whether we are using AI to diagnose lung cancer or machine learning to predict hospitalization risk, algorithms are transforming medicine and health care. Most of these are based on routinely collected health care data. But too often algorithms are deployed without a full understanding of what we are trying to predict and where the data are coming from. Through a combination of lectures, case studies, and interviews with leaders from across the health care ecosystem, we will discuss how to turn routine health care data into an algorithm, how to evaluate the validity of AI products, and how to understand liability and policy implications around algorithms in health care. This course will help you gain a fuller understanding of how to use data to transform care delivery and work with technical experts to design algorithms to fit your needs.

Course Topics
  • Big data
  • Machine learning
  • Predictive modeling
  • Human-algorithm collaboration
  • Bias in predictive analytics
  • Regulation, liability, and reimbursement of artificial intelligence


Introductory video to Using Data for Transformation

Learning Objectives

After completing this course, you will be able to:

  • Identify important sources of health care data and their limitations. 
  • Determine what types of challenges can be solved through the application of different types of algorithms, including those designed for prediction, diagnosis, estimation, and causal inference. 
  • Explain how to train and validate a predictive model.
  • Explain how to evaluate and mitigate bias in algorithms. 
  • Evaluate the legal and ethical implications of using artificial intelligence in health care settings. 
  • Design strategies to apply predictive models to maximize impact in health and health care.