Enhancing Extubation Failure Prediction with LLM-Derived Features from Respiratory Therapy Clinical Notes

利用呼吸治疗临床笔记中大语言模型提取的特征增强拔管失败预测

Abstract: Invasive mechanical ventilation is a lifesaving therapy, but timely, safe discontinuation is essential to preventing extubation failure (EF) and related risks to health. 摘要： 有创机械通气是一种挽救生命的治疗手段，但及时、安全地撤机对于预防拔管失败（EF）及相关健康风险至关重要。

We present a novel approach to EF prediction that leverages features classified in free-text respiratory therapy notes using a large language model and logistic regression pipeline. 我们提出了一种新的拔管失败预测方法，该方法利用大语言模型和逻辑回归流水线，对自由文本形式的呼吸治疗笔记中的特征进行分类。

Applied to a patient cohort from University of Washington Medicine, our method identifies clinically meaningful EF-related features that improve EF prediction performance when included alongside structured patient data. 通过应用于华盛顿大学医学中心（University of Washington Medicine）的患者队列，我们的方法识别出了具有临床意义的拔管失败相关特征；当这些特征与结构化患者数据结合使用时，能够显著提升拔管失败的预测性能。

We further highlight how differences in target populations in prior EF prediction studies, such as heterogenous inclusion criteria and EF definition, can lead to systematic differences in model performance and hinder generalizability between studies. 我们进一步强调了既往拔管失败预测研究中目标人群的差异（如纳入标准和拔管失败定义的异质性）如何导致模型性能的系统性差异，并阻碍了研究成果在不同场景间的泛化能力。