Artificial Intelligence for Pharmacoepidemiology Research: An Introduction
Synopsis
Recent rapid advances in the field of artificial intelligence (AI) will dramatically change many fields, including pharmacoepidemiology. This course will give an overview of foundational principles and concepts in machine learning (ML), AI, and natural language processing (NLP), and introduce their practical/potential applications in pharmacoepidemiology research.
Learning objectives
- Describe applications of AI/ML in pharmacoepidemiology research
- Recognize and differentiate different ML techniques
- Articulate the application of NLP on unstructured data to enhance pharmacoepidemiology studies
- Recognize bias/fairness and ethical issues of AI/ML application in pharmacoepidemiology.
- Distinguish AI/ML related prediction and description from causal inference
Course outline
- Overview of AI/ML
- Types of ML methods and algorithms (Yu Huang, University of Florida)
- The workflow of a ML project (Aokun Chen, University of Florida)
- NLP and unstructured data (Masoud Rouhizadeh, University of Florida)
- Deep learning (Aokun Chen, University of Florida)
- Bias, equity, and fairness assessment in AI/ML (Serena Jingchuan Guo, University of Florida)
- Causal inference in AI/ML (Tianze Jiao, University of Florida)
- Practical applications of AI/ML in pharmacoepidemiology research (Janick Weberpals, Harvard Medical School & Brigham and Women’s Hospital)
- Discussion and future directions