U.S. flag An official website of the United States government
  1. Home
  2. Science & Research
  3. About Science & Research at FDA
  4. FDA Grand Rounds
  5. FDA Grand Rounds – Adopting Large Language Models for Regulatory Review - 04/10/2025
  1. FDA Grand Rounds

Webcast | Virtual

Event Title
FDA Grand Rounds – Adopting Large Language Models for Regulatory Review
April 10, 2025

Date:
April 10, 2025
Time:
12:00 p.m. - 1:00 p.m. ET

Speaker 

Joshua Xu, Ph.D.

Joshua Xu, Ph.D.
Supervisory SBRBPAS Expert (Computer Scientist), R2R Branch Chief
Division of Bioinformatics and Biostatistics
Office of Research
National Center for Toxicological Research (NCTR)
Office of the Chief Scientist (OCS)
Office of the Commissioner (OC)

About the Speaker

Dr. Joshua Xu is the Branch Chief of Research-to-Review and Return (R2R) at the Division of Bioinformatics and Biostatistics within FDA’s National Center for Toxicological Research (NCTR). He has been (1) leading a community wide effort to assess the analytical performance of oncology panel sequencing technologies and develop best practice guidelines, (2) developing and adopting Natural Language Processing (NLP) methods for regulatory document review, (3) developing image analysis algorithms for digital pathology and food contamination detection.

About the Presentation

Generative AI such as large language models (LLMs) has demonstrated tremendous capabilities for comprehension, reasoning, and information synthesis. As the FDA continues to explore the potential of Gen AI to enhance the product review process, our research efforts have focused on evaluating and adapting locally hosted LLMs to a diverse set of regulatory review tasks. By going through several use cases including literature screening, adverse reaction detection from drug labeling, FDA Adverse Event Reporting System FAERS report deduplication, and AskFDADocuments – an AI chatbot for any FDA documents, this presentation will demonstrate the capabilities of LLMs and their potentials to support regulatory review process.

Learning Objectives

  1. Discuss the context-of-use and its importance in performance evaluation of AI application.
  2. Discuss the importance of quality data in the adoption of LLMs in regulatory review..

To Register

FDA employees MUST register in BOTH the following: 

Non-FDA employees must register in:

After MS Teams Webinar registration, you will receive a link via email to access the live webinar. You must log in with the username and password created when you registered.  Please pre-register at least two days before the event to ensure you receive the access link and outlook invitation for the session. 

For technical assistance please contact: Rakesh.Raghuwanshi@fda.hhs.gov
 

Back to Top