Linear Methods for Optimization and Prediction in Healthcare

Linear methods have traditionally been the workhorse of data analysis in many domains, and health-related applications are no exception. However, linear methods have a lot more to offer than standard regression analysis. This video explains why linear thinking remains a powerful and sophisticated way to think about data for prediction, causal analysis, and optimization in health tech. Designed for data scientists and for data savvy health care managers and clinicians, it demonstrates how to strengthen the conclusions you draw from health-related data and how to better allocate your health care resources. Examples are demonstrated in R and Python. Learners should understand the statistical limitations on extrapolating about groups and about the future based on limited sample size. They should also understand what it means to overfit a model and how to avoid doing so.

• Understand how to make causal inferences in health data using R and Python
• Explore techniques for assessing the strength of those causal inferences
• Discover methods for predicting population-level health parameters and individual outcomes
• Learn how to apply linear methods for health-related resource allocations

Aileen Nielsen is a software engineer at One Drop, an AI/ML health tech company working on diabetes management products. A member of the New York City Bar Association’s Science and Law committee, Aileen holds degrees in anthropology, law, and physics from Princeton, Yale, and Columbia, respectively. She focuses on improving daily life for underserved populations—particularly groups who have yet to fully enjoy the benefits of mobile technology.