AKIBoards: A Structure-Following Multiagent System for Predicting Acute Kidney Injury

2025-04-29

AI Summary

Addressing the challenges of collaborative multi-expert modeling and dynamic alignment between individual clinical reasoning and collective consensus in early acute kidney injury (AKI) prediction, this paper proposes STRUC-MAS—a Structured, Trust-Regulated, Uncertainty-Aware Collaborative Multi-Agent System. STRUC-MAS pioneers the explicit modeling of learnable, global medical knowledge structures as prior beliefs, embedded within an architecture-guided multi-agent system named AKIBoards. The framework integrates structured knowledge distillation, multi-agent fine-tuning, retrieval-augmented generation (RAG), and a confidence-driven interactive voting mechanism. Evaluated on 48-hour-ahead AKI prediction, the structure-adherent model achieves a mean average precision (AP) of 0.195—significantly surpassing the baseline (0.141). Agent interactions effectively enhance true-positive confidence and recalibrate belief distributions of low-recall agents, thereby enabling dynamic synergy between individual inference and group-level consensus.

Technology Category

• Multiagent Systems: Modeling other Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Knowledge Representation and Reasoning: Knowledge Acquisition

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Diagnostic reasoning entails a physician's local (mental) model based on an assumed or known shared perspective (global model) to explain patient observations with evidence assigned towards a clinical assessment. But in several (complex) medical situations, multiple experts work together as a team to optimize health evaluation and decision-making by leveraging different perspectives. Such consensus-driven reasoning reflects individual knowledge contributing toward a broader perspective on the patient. In this light, we introduce STRUCture-following for Multiagent Systems (STRUC-MAS), a framework automating the learning of these global models and their incorporation as prior beliefs for agents in multiagent systems (MAS) to follow. We demonstrate proof of concept with a prosocial MAS application for predicting acute kidney injuries (AKIs). In this case, we found that incorporating a global structure enabled multiple agents to achieve better performance (average precision, AP) in predicting AKI 48 hours before onset (structure-following-fine-tuned, SF-FT, AP=0.195; SF-FT-retrieval-augmented generation, SF-FT-RAG, AP=0.194) vs. baseline (non-structure-following-FT, NSF-FT, AP=0.141; NSF-FT-RAG, AP=0.180) for balanced precision-weighted-recall-weighted voting. Markedly, SF-FT agents with higher recall scores reported lower confidence levels in the initial round on true positive and false negative cases. But after explicit interactions, their confidence in their decisions increased (suggesting reinforced belief). In contrast, the SF-FT agent with the lowest recall decreased its confidence in true positive and false negative cases (suggesting a new belief). This approach suggests that learning and leveraging global structures in MAS is necessary prior to achieving competitive classification and diagnostic reasoning performance.

Problem

• Automating global model learning for multiagent diagnostic systems
• Improving acute kidney injury prediction via structure-following MAS
• Enhancing agent confidence through consensus-driven reasoning interactions

Innovation

• Automates learning of global models for MAS
• Uses structure-following for multiagent system
• Improves AKI prediction with prior beliefs