DIA 2025 Global Annual Meeting

Component Type: Session
Level: Intermediate
CE: ACPE 1.00 Knowledge UAN: 0286-0000-25-580-L04-P; CME 1.00; RN 1.00

As the potential of artificial intelligence (AI) is increasingly piquing the interest of researchers, it is critical to discuss how AI works by learning from the data, and how to leverage the results. Choosing the right approach for integrating AI into the research lifecycle can have a massive impact on likelihood of success. As one example, both generalized large language models (LLMs) and purpose build LLMs have their place in the toolkit needed to curate and analyze medical record and other patient data. This presentation will explore best practices/learnings for ensuring the right AI tool for the task, including LLMs tailored to the unique needs of pharmaceutical research; the impact for pharmaceutical research and for patients this can have; and imperatives for ascertaining the reliability and trustworthiness of AI solutions for medical research, as key to promoting adoption and responsible deployment in practice.

Learning Objectives

Identify key factors to consider when selecting AI strategies and methods for use in biomedical settings and pharmaceutical research; Describe the potential impact of AI strategies on research quality and patient outcomes, using real-world use cases and examples; Apply a framework to evaluate the suitability and readiness of AI technologies, such as large language models, for specific research or clinical applications.

Chair

Sarah Lyons, MSc

Speaker

Not all AI is Equal: Ensuring Reliable, Responsible AI to Accelerate Adoption, Innovation, and Impact
Alex Asiimwe

Deploying Large Language Models in Clinical Research: Practical Considerations from Real World Examples
Troy Astorino