Professor and Chair Department of Biomedical Informatics, College of Medicine, The Ohio State University Columbus, Ohio, United States
Disclosure information not submitted.
This presentation will demonstrate how ML/AI and human annotators can iteratively refine natural language processing (NLP) analysis to identify published pharmacokinetics studies (PK), clinical trials (CT), and pharmaco-epidemiology studies (PE) in PubMed using large language models and active learning strategies. The focus will be on pharmacotherapies in pediatric and maternal patient populations. Their drug efficacy and side effect data rely on these PK/CT/PE studies. Because of excellent NLP performance of large language models, we will identify knowledge gaps among all the drugs in both pediatric and maternal patients.
