Division Director US Food and Drug Administration Silver Spring, Maryland
This presentation will illuminate the potential of specialized AI and GPT tools in revolutionizing the landscape of drug development and regulatory assessments. Through the exploration of these case examples, we aim to demonstrate how these innovative technologies can not only streamline processes but also contribute to the generative nature of AI tools, ultimately benefiting the healthcare industry and patients alike.
In the first case example, we will examine the effective utilization of machine learning techniques in predicting safety profiles for kinase inhibitors during the early stages of drug discovery. This showcases the potential of AI to significantly expedite the drug development process while ensuring a heightened level of safety assessment. By leveraging machine learning, we can analyze vast datasets and extract valuable insights, ultimately reducing the time and resources required for the development of new pharmaceuticals.
The second case example underscores best practices in the development of domain-specific GPT models designed to process and comprehend intricate drug labeling information. These models are capable of generating text that aligns with the requirements of regulatory assessors. This represents a crucial advancement in simplifying the regulatory assessment process, as the generation of precise, assessor-desired texts can greatly enhance the efficiency and accuracy of regulatory compliance.
Learning Objectives:
Upon completion, participants will be able to have a preliminary understanding of current use of machine learning in drug discovery
Upon completion, participants will be able to understand the current best practice to develope domain specific GPT models for regulatory use
Upon completion, participants will be able to understand potential future directions of GPT use in drug development