Organ-Agents: Virtual Human Physiology Simulator via LLMs

2025-08-19

AI Summary

To address the insufficient trustworthiness and interpretability of digital twin systems in precision medicine, this study proposes a large language model (LLM)-driven multi-agent physiological simulation framework. The framework models nine core physiological systems—including cardiovascular, renal, and immune—and integrates temporal clinical data via supervised fine-tuning. It introduces a reinforcement-based coordination mechanism featuring dynamic reference selection and error correction to enable collaborative multi-agent reasoning and counterfactual treatment simulation. The method supports physiological plausibility verification and hypothesis-driven inference. Evaluated on over 10,000 patient cases, the framework achieves mean squared errors <0.16 across all physiological systems, attains an average clinician-assessed score of 3.9/5, and incurs <0.04 AUROC degradation for classifiers trained on synthetic data—demonstrating strong clinical applicability and generalization capability.

Technology Category

• Multiagent Systems: Modeling other Agents
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Recent advances in large language models (LLMs) have enabled new possibilities in simulating complex physiological systems. We introduce Organ-Agents, a multi-agent framework that simulates human physiology via LLM-driven agents. Each Simulator models a specific system (e.g., cardiovascular, renal, immune). Training consists of supervised fine-tuning on system-specific time-series data, followed by reinforcement-guided coordination using dynamic reference selection and error correction. We curated data from 7,134 sepsis patients and 7,895 controls, generating high-resolution trajectories across 9 systems and 125 variables. Organ-Agents achieved high simulation accuracy on 4,509 held-out patients, with per-system MSEs <0.16 and robustness across SOFA-based severity strata. External validation on 22,689 ICU patients from two hospitals showed moderate degradation under distribution shifts with stable simulation. Organ-Agents faithfully reproduces critical multi-system events (e.g., hypotension, hyperlactatemia, hypoxemia) with coherent timing and phase progression. Evaluation by 15 critical care physicians confirmed realism and physiological plausibility (mean Likert ratings 3.9 and 3.7). Organ-Agents also enables counterfactual simulations under alternative sepsis treatment strategies, generating trajectories and APACHE II scores aligned with matched real-world patients. In downstream early warning tasks, classifiers trained on synthetic data showed minimal AUROC drops (<0.04), indicating preserved decision-relevant patterns. These results position Organ-Agents as a credible, interpretable, and generalizable digital twin for precision diagnosis, treatment simulation, and hypothesis testing in critical care.

Problem

• Simulating human physiological systems using LLM-driven multi-agent framework
• Achieving high accuracy in patient physiology simulation across multiple systems
• Enabling counterfactual treatment simulations and early warning tasks

Innovation

• Multi-agent LLM framework simulates physiology
• Supervised fine-tuning with reinforcement-guided coordination
• Generates synthetic data for clinical applications