COPE：基于开源大语言模型的链式推理预测引擎用于从临床笔记预测中风结局
COPE: Chain-Of-Thought Prediction Engine for Open-Source Large Language Model Based Stroke Outcome Prediction from Clinical Notes
AI总结：
COPE通过链式推理框架从临床笔记预测中风预后，优于现有模型，提供轻量级且可解释的解决方案。
摘要：
预测急性缺血性中风（AIS）的预后有助于临床决策、患者咨询和资源分配。临床笔记包含丰富的上下文信息，但其非结构化性质限制了其在传统预测模型中的使用。我们开发并评估了链式推理（CoT）结局预测引擎（COPE），这是一种增强推理能力的大型语言模型框架，用于从非结构化临床笔记中预测AIS患者90天的功能预后。本研究包括464名AIS患者，其出院小结和90天修改Rankin评分（mRS）评分。COPE使用基于顺序开源LLaMA-3-8B模型的两步CoT框架：第一步生成临床推理，第二步输出mRS预测。我们比较了COPE与GPT-4.1、ClinicalBERT、基于结构化变量的机器学习模型（Clinical ML）以及无CoT的单步LLM。性能通过均方误差（MAE）、±1 mRS点内的准确率和精确准确率进行评估。COPE的MAE为1.01（95%置信区间0.92-1.11），±1点内的准确率为74.4%（69.9, 78.8%），精确准确率为32.8%（28.0, 37.6%），与GPT-4.1相当，优于ClinicalBERT [MAE 1.24（1.13-1.36）]、Clinical ML [1.28（1.18-1.39）]和单步LLM [1.20（1.09-1.33）]。亚组分析显示，性能在性别和年龄方面保持一致，但老年人、接受取栓术的患者以及摘要较长的患者误差略高。这些发现表明，COPE是一种轻量级、可解释且隐私保护的开源框架，为从非结构化临床文本中进行预后预测提供了准确且实用的解决方案。
英文摘要：
Predicting outcomes in acute ischemic stroke (AIS) guides clinical decision-making, patient counseling, and resource allocation. Clinical notes contain rich contextual information, but their unstructured nature limits their use in traditional predictive models. We developed and evaluated the Chain-of-Thought (CoT) Outcome Prediction Engine (COPE), a reasoning-enhanced large language model framework, for predicting 90-day functional outcomes after AIS from unstructured clinical notes. This study included 464 AIS patients with discharge summaries and 90-day modified Rankin Scale (mRS) scores. COPE uses a two-step CoT framework based on sequential open-source LLaMA-3-8B models: the first generates clinical reasoning, and the second outputs an mRS prediction. We compared COPE with GPT-4.1, ClinicalBERT, a structured variable-based machine learning model (Clinical ML), and a single-step LLM without CoT. Performance was evaluated using mean absolute error (MAE), accuracy within +/-1 mRS point, and exact accuracy. COPE achieved an MAE of 1.01 (95% CI 0.92-1.11), +/-1 accuracy of 74.4% (69.9, 78.8%), and exact accuracy of 32.8% (28.0, 37.6%), comparable to GPT-4.1 and superior to ClinicalBERT [MAE 1.24 (1.13-1.36)], Clinical ML [1.28 (1.18-1.39)], and the single-step LLM [1.20 (1.09-1.33)]. Subgroup analyses showed consistent performance across sex and age, with slightly higher error among older patients, those undergoing thrombectomy, and those with longer summaries. These findings demonstrate that COPE, a lightweight, interpretable, and privacy-preserving open-source framework, provides an accurate and practical solution for outcome prediction from unstructured clinical text.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
