HERMES: Contrast-Aware Knowledge Graph Reasoning from Clinical Notes for Patient Outcome Prediction

HERMES：基于临床笔记的对比感知知识图谱推理用于患者预后预测

Abstract: Clinical predictive models often rely on structured Electronic Health Record data, such as time-series and procedure codes. While recent approaches have begun leveraging unstructured clinical notes, they typically encode them as flat sequences, which may lose explicit relational and temporal structure present in clinical narratives.

摘要： 临床预测模型通常依赖于结构化的电子健康记录（EHR）数据，例如时间序列和程序代码。尽管近期的一些方法已开始利用非结构化的临床笔记，但它们通常将其编码为扁平序列，这可能会丢失临床叙述中存在的显式关系和时间结构。

In response, we propose HERMES, a graph-based framework that operates exclusively on clinical text while preserving clinical relationships. This approach builds on two key ideas. First, personalized Knowledge Graphs (KGs) are constructed through Large-Language-Model-guided extraction from clinical notes with Contrastive Logic Modeling that explicitly captures temporal dynamics and treatment failures and changes in outcomes.

为此，我们提出了 HERMES，这是一个基于图的框架，它仅在临床文本上运行，同时保留了临床关系。该方法基于两个关键理念。首先，通过大语言模型引导从临床笔记中提取信息，并结合对比逻辑建模（Contrastive Logic Modeling）构建个性化知识图谱（KG），从而显式地捕捉时间动态、治疗失败以及预后的变化。

Second, a Graph Attention Network synthesizes patient representations through graph-based learning over the KGs. Experiments on MIMIC-III and MIMIC-IV for in-hospital mortality and 30-day readmission prediction show that HERMES consistently outperforms strong text-only baselines. Our findings demonstrate that explicit relational modeling with Contrastive Logic Modeling significantly advances predictive performance.

其次，图注意力网络（Graph Attention Network）通过对知识图谱进行基于图的学习来综合患者表征。在 MIMIC-III 和 MIMIC-IV 数据集上进行的院内死亡率和 30 天再入院预测实验表明，HERMES 的表现始终优于强大的纯文本基线模型。我们的研究结果证明，结合对比逻辑建模的显式关系建模显著提升了预测性能。