Agentic AI for Laboratory Analytics

Nicholas C. Spies (1,2), Paul English (2), Brendan O'Fallon (2), David Ng (1,2)
(1) Department of Pathology, University of Utah Health, Salt Lake City, UT, USA. (2) ARUP Laboratories, Salt Lake City, UT, USA.

Nicholas Spies, MD (Presenter) University of Utah, ARUP Laboratories

Presenter Bio: Nick Spies, MD, is a bioinformatician-turned-laboratorian who is a medical director in the Applied Artificial Intelligence group within ARUP laboratories' Division of Research and Innovation. He is focused on applying analytical techniques to improve the way we detect laboratory errors, and hopes to spread the good word of data science and machine learning within the laboratory medicine community.

Relevant Financial Disclosures (within past 24 months, reported on Apr 30, 2026) No relevant financial relationship(s) to disclose.

Generative AI applications have recently expanded beyond passive information retrieval to autonomous, goal-oriented “agents" capable of navigating complex data analytics tasks. These agents can use tools such as python or SQL to analyze data, and can function within secure, sandboxed environments isolated behind enterprise firewalls to generate insights from proprietary or protected data. This talk will explore the experience of ARUP Laboratories' Applied AI group as they develop, validate, and deploy an agentic AI solution for laboratory directors to accomplish data analytics tasks for clinical research and operations.