A two-agent structure for enhanced reasoning
Our work addresses this problem with a novel strategy based mostly on the interaction of two LLM-driven brokers, which is analogous to how human clinicians sort out administration issues.
The Dialogue Agent is user-facing and geared up to quickly reply based mostly on its present understanding of the affected person. This agent handles the conversational facets of the interplay, gathering details about the affected person’s situation, addressing their issues, and constructing rapport. By leveraging pure language processing and empathetic communication methods, the Dialogue Agent ensures a seamless and interesting consumer expertise.
The Mx Agent (Administration Reasoning Agent) intentionally and repeatedly analyzes the accessible info, together with medical pointers and patient-specific information, to optimize administration of the affected person. Leveraging Gemini’s state-of-the-art long-context capabilities, this agent synthesizes and causes over massive quantities of data — affected person dialogues throughout a number of visits along with lots of of pages of medical pointers — suddenly. Utilizing this strategy, it produces structured plans for investigations, remedies, and follow-up care, bearing in mind the most recent medical proof, info gathered throughout earlier visits, and particular person affected person preferences.
