Leveraging Natural Language Processing, Machine Learning, and Social Determinants in Electronic Health Records to Improve Opioid Use Disorder Clinical Care

2025 Informatics Summit On Demand

Presentation Time: 03:30 PM - 05:00 PM

Abstract Keywords: Clinical Decision Support for Translational/Data Science Interventions, Natural Language Processing, Machine Learning, Generative AI, and Predictive Modeling, Social Determinants of Health
Primary Track: Data Science/Artificial Intelligence
Programmatic Theme: Health Data Science and Artificial Intelligence Innovation: From Single-Center to Multi-Site

Advances in artificial intelligence (AI) approaches such as natural language processing (NLP) and machine learning (ML) have the potential to transform clinical care but come with inherent risks. In the context of the ongoing opioid overdose crisis, health systems are struggling to identify patients with opioid use disorder and address social and behavioral factors that influence care provision. AI holds promise for increasing the accuracy and efficiency of patient identification and care delivery. Investigators will discuss methods for ascertaining social determinants associated with overdose in unstructured electronic health record data using NLP, identifying patients with opioid misuse through NLP and ML algorithms in the emergency department, and integrating social risk factors into clinical decision support to promote enhanced treatment for opioid use disorder with medications. Benefits and risks will be discussed, with a focus on trustworthy AI. NIH will share scientific priorities and funding opportunities to support such research.
Moderator:
Tamara Haegerich, PhD
National Institutes of Health

Speaker(s):
Yu Hong, MS, PhD, FACMI
University of Massachusetts Lowell, VA Bedford Healthcare

Melanie Molina, MD, MAS
University of California San Francisco

Neeraj Chhabra, MD
University of Illinois Chicago