Standalone LLM and a Pre-specified Agentic Pipeline for Explaining ICU Mortality Predictions: a Feasibility Study on the eICU Demo Dataset

独立大语言模型与预设智能体工作流在ICU死亡率预测解释中的应用：基于eICU演示数据集的可行性研究

Abstract: Machine-learning models can predict ICU mortality accurately, but feature-attribution methods alone rarely provide the clinical narrative needed for bedside use. Large language models (LLMs) may bridge this gap, and multi-step agentic pipelines are a plausible extension because they separate data interpretation, guideline checking, and final explanation. 摘要： 机器学习模型能够准确预测ICU（重症监护室）的死亡率，但仅靠特征归因方法往往无法提供临床床旁应用所需的临床叙述。大语言模型（LLMs）或许能弥补这一差距，而多步智能体工作流（Agentic pipelines）作为一种合理的扩展，能够将数据解读、指南核查和最终解释过程分离开来。

This revised feasibility study preserves the original standalone-versus-agentic comparison while making the main clinical findings more explicit. Using the retained local eICU Demo artifact set (2,353 ICU stays; 8.1% mortality), XGBoost achieved an AUROC of 0.855 (95% CI 0.796—0.906) and an AUPRC of 0.332 (95% CI 0.217—0.494). 这项修订后的可行性研究保留了最初的“独立模型与智能体工作流”对比，并使主要的临床发现更加明确。使用保留的本地eICU演示数据集（包含2,353次ICU住院记录；死亡率为8.1%），XGBoost模型实现了0.855的AUROC（95%置信区间为0.796—0.906）和0.332的AUPRC（95%置信区间为0.217—0.494）。

On a stratified 38-case explanation subset, the standalone LLM produced 1 explanation with explicit outcome leakage, whereas the four-step agentic pipeline produced none. Among the 14 cases that overlapped with the SHAP review subset, the standalone LLM showed higher SHAP alignment (mean Jaccard 0.171 versus 0.077) and higher direction consistency (92.9% versus 78.6%), while the agentic pipeline showed higher guideline grounding (0.762 versus 0.143), higher value specificity (0.236 versus 0.143), and slightly higher plausibility (0.700 versus 0.671). 在分层的38例解释子集中，独立大语言模型产生了一例存在明显结果泄露的解释，而四步智能体工作流则未出现此类情况。在与SHAP审查子集重叠的14个案例中，独立大语言模型表现出更高的SHAP对齐度（平均Jaccard系数为0.171对0.077）和更高的一致性（92.9%对78.6%）；而智能体工作流则表现出更高的指南依据性（0.762对0.143）、更高的数值特异性（0.236对0.143）以及略高的合理性（0.700对0.671）。

Clinically, the results suggest that agentic decomposition may improve safety-relevant grounding and patient-specific detail, but it should be paired with attribution-based checks before use in high-stakes risk explanation. 从临床角度来看，研究结果表明，智能体分解方法可以提高与安全性相关的依据性和患者特异性细节，但在用于高风险解释场景之前，应将其与基于归因的检查相结合。