ReToP:学习重写电子健康记录以进行临床预测
ReToP: Learning to Rewrite Electronic Health Records for Clinical Prediction
AI总结:
ReToP通过端到端训练EHR重写器和临床预测器,利用分类监督贡献分数提升临床预测性能,实现EHR数据的优化重写与预测任务的结合。
摘要:
电子健康记录(EHRs)为临床决策提供了关键信息。然而,其高维性、异质性和稀疏性使临床预测变得具有挑战性。大型语言模型(LLMs)通过利用参数化医学知识来增强EHR数据,为解决这一挑战带来了进展。尽管迄今为止取得了显著成就,但大多数现有方法本质上是任务无关的,即它们将LLMs用作EHR编码器或EHR完成模块,而没有完全整合来自预测任务的信号。这自然地阻碍了任务性能的准确性。在本工作中,我们提出了Rewrite-To-Predict(ReToP),一个基于LLM的框架,通过端到端训练EHR重写器和临床预测器来解决这一限制。为应对EHR重写训练数据不足的问题,我们使用临床驱动的特征选择策略生成合成伪标签,以创建多样化的患者重写内容,用于微调EHR重写器。ReToP通过一种新的分类监督贡献(CSC)分数将重写器与预测目标对齐,使EHR重写器能够生成具有临床相关性的重写内容,从而直接增强预测。我们的ReToP框架在MIMIC-IV的三个临床任务上超过了强大的基线模型。此外,ReToP的分析显示其在未见过的数据集和任务上具有良好的泛化能力,且在最小微调的情况下仍能保持忠实的重写和强调任务相关的预测特征。
英文摘要:
Electronic Health Records (EHRs) provide crucial information for clinical decision-making. However, their high-dimensionality, heterogeneity, and sparsity make clinical prediction challenging. Large Language Models (LLMs) allowed progress towards addressing this challenge by leveraging parametric medical knowledge to enhance EHR data for clinical prediction tasks. Despite the significant achievements made so far, most of the existing approaches are fundamentally task-agnostic in the sense that they deploy LLMs as EHR encoders or EHR completion modules without fully integrating signals from the prediction tasks. This naturally hinders task performance accuracy. In this work, we propose Rewrite-To-Predict (ReToP), an LLM-based framework that addresses this limitation through an end-to-end training of an EHR rewriter and a clinical predictor. To cope with the lack of EHR rewrite training data, we generate synthetic pseudo-labels using clinical-driven feature selection strategies to create diverse patient rewrites for fine-tuning the EHR rewriter. ReToP aligns the rewriter with prediction objectives using a novel Classifier Supervised Contribution (CSC) score that enables the EHR rewriter to generate clinically relevant rewrites that directly enhance prediction. Our ReToP framework surpasses strong baseline models across three clinical tasks on MIMIC-IV. Moreover, the analysis of ReToP shows its generalizability to unseen datasets and tasks with minimal fine-tuning while preserving faithful rewrites and emphasizing task-relevant predictive features.
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。