Breast cancer diagnosis and treatment planning demands precision across a complex workflow involving disparate healthcare systems and multidisciplinary teams. This session explores how agentic AI is transforming oncology care through autonomous decision-making capabilities and adaptive learning in complex healthcare environments, with particular emphasis on enhancing the patient experience across fragmented systems. We will explore how agentic AI addresses the aggregation and interpretation of multimodal data from imaging studies, pathology reports, genomic sequencing, clinical histories, and real-time monitoring. 

The session demonstrates how multi-agent solutions employ distributed problem-solving approaches to process different aspects of complex clinical data while coordinating care across multiple touchpoints, creating a unified view of the patient’s condition. Through real-world examples, we will illustrate how the orchestration design pattern enables asynchronous workflow management, dynamically coordinating complex tasks by intelligently planning, decomposing, and delegating subtasks to specialized worker agents. The presentation emphasizes the critical importance of human-in-the-loop design, ensuring AI serves as a powerful augmentation tool rather than a replacement for clinical expertise.

Learning Objectives

• Evaluate how agentic AI integrates multimodal data, improving diagnostic precision
• Design a multi-agent collaborative workflow by mapping distributed problem-solving approaches when coordinating oncology care across multiple touchpoints
• Implement “human-in-the-loop” by configuring AI task delegation and decomposition strategies that correctly maintain clinical expertise as the primary decision authority