arXiv 2509.188462025-09-23cs.AI

模型选择遇上临床语义：通过LLM-as-Judge评估、冗余感知采样和分段感知微调优化ICD-10-CM预测

Model selection meets clinical semantics: Optimizing ICD-10-CM prediction via LLM-as-Judge evaluation, redundancy-aware sampling, and section-aware fine-tuning

National Kaohsiung University of Science and Technology（高雄科技大学）；National Health Research Institutes（国家卫生研究院）；Kaohsiung Medical University（高雄医学大学）；Taitung MacKay Memorial Hospital（台东马偕纪念医院）；MacKay Medical University（马偕医科大学）；Kaohsiung Medical University Hospital（高雄医学大学附设医院）

Hong-Jie Dai, Zheng-Hao Li, An-Tai Lu, Bo-Tsz Shain, Ming-Ta Li, Tatheer Hussain Mir, Kuang-Te Wang, Min-I Su, Pei-Kang Liu, Ming-Ju Tsai

备注：: 28 Pages, 4 Figures, 2 Tables

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出一个模块化框架，通过LLM-as-Judge评估与Plackett-Luce聚合选择最优基础模型，结合冗余感知采样和分段感知微调，在台湾医院数据上显著提升了ICD-10-CM代码预测性能。

摘要：

准确的国际疾病分类（ICD）编码对于临床记录、计费和医疗分析至关重要，但它仍然是一项劳动密集且容易出错的任务。尽管大型语言模型（LLMs）在自动化ICD编码方面显示出前景，但它们在基础模型选择、输入情境化和训练数据冗余方面的挑战限制了其有效性。我们提出了一个用于ICD-10临床修改版（ICD-10-CM）代码预测的模块化框架，通过有原则的模型选择、冗余感知的数据采样和结构化输入设计来应对这些挑战。该框架集成了LLM-as-Judge评估协议与Plackett-Luce聚合方法，根据开源LLMs对ICD-10-CM代码定义的内在理解能力对其进行评估和排序。我们引入了基于嵌入的相似性度量，以及一种冗余感知采样策略，用于去除语义重复的出院摘要。我们利用来自台湾医院的结构化出院摘要来评估上下文效果，并在通用和分段特定建模范式下考察分段内容纳入的影响。在两个机构数据集上的实验表明，经过微调后选定的基础模型在内部和外部评估中始终优于基线LLMs。纳入更多临床分段持续改善了预测性能。本研究使用开源LLMs建立了一种实用且有原则的ICD-10-CM代码预测方法。所提出的框架通过结合明智的模型选择、高效的数据精炼和上下文感知提示，为自动化医疗编码系统的实际部署提供了一种可扩展的、机构就绪的解决方案。

英文摘要：

Accurate International Classification of Diseases (ICD) coding is critical for clinical documentation, billing, and healthcare analytics, yet it remains a labour-intensive and error-prone task. Although large language models (LLMs) show promise in automating ICD coding, their challenges in base model selection, input contextualization, and training data redundancy limit their effectiveness. We propose a modular framework for ICD-10 Clinical Modification (ICD-10-CM) code prediction that addresses these challenges through principled model selection, redundancy-aware data sampling, and structured input design. The framework integrates an LLM-as-judge evaluation protocol with Plackett-Luce aggregation to assess and rank open-source LLMs based on their intrinsic comprehension of ICD-10-CM code definitions. We introduced embedding-based similarity measures, a redundancy-aware sampling strategy to remove semantically duplicated discharge summaries. We leverage structured discharge summaries from Taiwanese hospitals to evaluate contextual effects and examine section-wise content inclusion under universal and section-specific modelling paradigms. Experiments across two institutional datasets demonstrate that the selected base model after fine-tuning consistently outperforms baseline LLMs in internal and external evaluations. Incorporating more clinical sections consistently improves prediction performance. This study uses open-source LLMs to establish a practical and principled approach to ICD-10-CM code prediction. The proposed framework provides a scalable, institution-ready solution for real-world deployment of automated medical coding systems by combining informed model selection, efficient data refinement, and context-aware prompting.

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。