Beyond the Individual: Virtualizing Multi-Disciplinary Reasoning for Clinical Intake via Collaborative Agents

2026-04-23 · Source: cs.MA updates on arXiv.org · Field: Health & Wellbeing — Medical Devices & Health Technology, Clinical Care & Medical Practice, Health & Medical Research · Depth: Expert, extended

Summary

Aegle is a novel multi-agent framework designed to virtualize Multi-Disciplinary Team (MDT) reasoning for clinical intake, addressing the limitations of single-physician consultations. It employs a graph-based architecture with an orchestrator, specialist agents, and an aggregator, formalizing the consultation state using a structured SOAP (Subjective, Objective, Assessment, Plan) representation. This framework separates evidence collection from diagnostic reasoning to enhance traceability and mitigate cognitive biases. Aegle dynamically activates specialist agents for decoupled parallel reasoning, integrating their outputs into a coherent clinical note. Evaluated on ClinicalBench and a real-world RAPID-IPN dataset across 24 departments and 53 metrics, Aegle consistently outperforms state-of-the-art proprietary and open-source models in documentation quality, consultation capability, and final diagnosis accuracy. The code is publicly available at https://github.com/HovChen/Aegle.

Key takeaway

For AI Scientists and Machine Learning Engineers developing clinical decision support systems, Aegle's virtual MDT framework offers a robust approach to mitigate cognitive biases and improve diagnostic accuracy during patient intake. You should consider adopting a structured, multi-agent architecture with dynamic specialist activation and explicit separation of evidence collection from diagnostic synthesis to enhance the reliability and scalability of your clinical AI assistants.

Key insights

Aegle virtualizes MDT reasoning for clinical intake using a multi-agent framework, improving diagnostic accuracy and documentation quality.

Principles

• Decoupled parallel reasoning enhances hypothesis diversity.
• Structured clinical state (SOAP) improves traceability and bias control.
• Dynamic topology optimizes specialist activation efficiency.

Method

Aegle uses an Orchestrator to dynamically activate specialist agents for parallel reasoning based on a structured SOAP state, followed by an Aggregator that integrates outputs into a coherent clinical note through a write-then-speak protocol.

In practice

• Implement structured SOAP state for evidence-first reasoning.
• Utilize dynamic agent activation to reduce inference overhead.
• Separate evidence acquisition from diagnostic synthesis.

Topics

• Aegle Framework
• Virtual MDT
• Multi-Agent Systems
• Clinical Intake
• SOAP Documentation

Code references

• HovChen/Aegle

Best for: AI Scientist, Research Scientist, Machine Learning Engineer

Related on AIssential

• MedRLM: Recursive Multimodal Health Intelligence for Long-Context Clinical Reasoning, Sensor-Guided Screening, Evidence-Grounded Decision Support, and Community-to-Tertiary Referral Optimization — MedRLM enables recursive, multimodal clinical reasoning by treating patient data as an inspectable external environment.
• Enhancing Clinical Trial Patient Matching through Knowledge Augmentation and Reasoning with Multi-Agent — Multi-agent systems can augment LLMs with domain knowledge to improve clinical trial patient matching accuracy.
• Treatment Response Optimized Clinical Decision Support AI System via Digital Twin Simulation — An online adaptive AI framework integrates TE, DT, and RL for safe, continuously improving clinical decision support.
• Exploring the feasibility of conversational diagnostic AI in a real-world clinical study — Supervised conversational AI for pre-visit history taking is feasible, safe, and well-received by patients and clinicians.
• Active Evidence-Seeking and Diagnostic Reasoning in Large Language Models for Clinical Decision Support — LLMs exhibit a substantial performance gap in active, multi-turn clinical diagnosis compared to static, full-context evaluations.
• Aether: Network Validation Using Agentic AI and Digital Twin — Aether automates network change validation using agentic AI and a multi-functional digital twin.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.MA updates on arXiv.org.