TRI01: Advanced AI Architectures & Human Factors: Ambient Agents, Evolutionary Discovery, and Safe Implementation in Healthcare (Workshop)
Presentation Type: Workshop
Not recorded for AMIA Now
Session Credits: 2
Advanced AI Architectures & Human Factors: Ambient Agents, Evolutionary Discovery, and Safe Implementation in Healthcare
Presentation Type: Workshop - Instructional
Not recorded for AMIA Now
Presentation Time: 08:00 AM - 10:00 AM
Primary Track: Clinical Research Informatics
This advanced instructional workshop targets informatics professionals ready to move beyond standard Large Language Model (LLM) deployments. The session explores the cutting edge of AI utility in healthcare through the intersection of technical architecture and human-centered design. Participants will examine three distinct pillars: Ambient Intelligence, Autonomous Agents, and Evolutionary Discovery. We will deconstruct the architecture of ambient listening technologies and couple them with agentic systems that can reason, plan, and execute clinical tasks. Crucially, we will apply Human Factors (HF) engineering principles to these architectures, exploring how to design "Human-AI Loops" that ensure safety, trust calibration, and explainability (XAI) in high-stakes clinical environments. Finally, the workshop introduces the frontier of "AI for Science," utilizing frameworks like AlphaEvolve to demonstrate how LLMs can autonomously discover optimization strategies. Through architectural deep-dives and case studies, attendees will learn to design self-improving systems that actively assist in problem-solving while minimizing cognitive load and maintaining provider trust.
