文献速递胸部技术与安全CT

大语言模型扩展RadLex本体的胸部CT报告多国数据验证

Large Language Model-Generated Expansion of the RadLex Ontology: Application to Multinational Datasets of Chest CT Reports.
PubMed 文献 · AJR Am J Roentgenol · 2026-01-14

本研究使用大语言模型对RadLex放射学术语本体进行词汇变体与同义词扩展,并在五组多国胸部CT报告数据集中评估扩展前后的词汇覆盖率和语义识别效能。结果显示,LLM生成扩展相较RadLex自带扩展,词汇覆盖率与召回率更高,精确度略降,F1值更高。该扩展方法为放射学本体扩充提供了可扩展的实用方案。

要点

  • LLM扩展RadLex可将胸部CT报告词汇覆盖率从67.5-75.4%提升至81.9-86.2%。
  • 语义召回率提升至81.6-91.4%,伴少量精确度下降(94.8-98.2%)。
  • 扩展词表可用于优化自然语言处理对放射报告的语义识别。

核心结论

  • 在多国胸部CT报告数据集中,LLM生成术语扩展较RadLex自带扩展提升词汇覆盖率与语义召回率,仅少量损失语义精确度。
  • 基于LLM的扩展方法能保持语义对齐,为放射学本体扩充提供实用且可扩展的解决方案。

研究设计与样本

  • 回顾性研究,使用大语言模型Gemini 2.0 Flash Thinking扩展RadLex约40000个首选术语。
  • 共纳入五组胸部CT报告数据集:韩国机构两组(n=119098、n=245),西班牙n=5213,土耳其n=21304,美国n=19402。
  • 同一LLM将报告解析为词法单元,自动计算各数据集的词汇覆盖率;另从每组随机抽取100份报告人工评估精确度、召回率与F1值。

诊断标准与关键数值

  • RadLex自带扩展含17515个术语;LLM生成扩展含208465个词汇变体及69918个同义词。
  • LLM扩展对全部五组数据集的词汇覆盖率为81.9-86.2%,RadLex自带扩展为67.5-75.4%。
  • LLM扩展的召回率为81.6-91.4%,高于RadLex自带扩展的64.0-80.3%。
  • LLM扩展的精确度为94.8-98.2%,低于RadLex自带扩展的100.0%。
  • LLM扩展的F1值为0.91-0.95,高于RadLex自带扩展的0.86-0.91。

临床应用与阅片要点

  • LLM生成的扩展词表可弥补放射报告语言变异导致的术语覆盖不足,利于真实世界自然语言处理应用。
  • 在自动识别报告语义单元时,需注意扩展词表可能导致一定程度的精确度下降。

局限性

  • 研究为回顾性设计,报告仅来自胸部CT领域,结论向其他部位或模态的推广性未在本文验证。
  • 精确度下降提示扩展词表可能引入少量语义不匹配,需在实际应用中加以权衡。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

BACKGROUND. RadLex (Radiological Society of North America) is a widely used radiology-specific ontology that standardizes terminology for clinical and research uses. However, the ontology's coverage of clinical radiology reports remains limited due to radiologists' linguistic variation. OBJECTIVE. The purpose of this study was to use a large language model (LLM) to generate an expanded set of lexical variants and synonyms for the RadLex ontology and to evaluate the impact of this expansion on lexical coverage and semantic term recognition using clinical radiology reports. METHODS. This retrospective study used an LLM (Gemini 2.0 Flash Thinking, Google) to generate an expansion (lexical variants [morphologic variants, orthographic variants, and acronyms and abbreviations] and strict semantic synonyms) of the 40,000 RadLex preferred terms, with detailed constraints to ensure semantic alignment. Five datasets of clinical chest CT reports were obtained (two from the study institution in Korea [n = 119,098 and n = 245] and three from public datasets [Spain, n = 5213; Turkey, n = 21,304; United States, n = 19,402]). The same LLM was used to parse the reports into lexicon units (concise text strings representing distinct medical concepts). For each dataset, the lexical coverage rate was automatically computed as a measure of the extent to which the parsed units matched a given expression list. Additionally, 100 randomly selected reports from each dataset were manually reviewed to determine a given expression list's precision, recall, and F1 score (measures of unit-level matching performance when requiring semantic fidelity). Metrics were compared between the existing RadLex-provided expansion and the LLM-generated expansion. RESULTS. The RadLex-provided expansion contained 17,515 terms. The LLM-generated expansion contained 208,465 lexical variants and 69,918 synonyms. For all five datasets, the LLM-generated expansion, compared with the RadLex-provided expansion, had a greater lexical coverage rate (81.9-86.2% vs 67.5-75.4%), greater recall (81.691.4% vs 64.0-80.3%), lower precision (94.8-98.2% vs 100.0% [all datasets]), and greater F1 score (0.91-0.95 vs 0.86-0.91). CONCLUSION. Across multinational datasets of clinical chest CT reports, the LLM-generated term expansion yielded improved lexical coverage and semantic recall, with only small loss of semantic precision, compared with the RadLex-provided expansion. CLINICAL IMPACT. The LLM-based approach provides a practical and scalable solution for expanding radiology ontologies while maintaining semantic alignment; the method can aid real-world natural language processing applications.

来源:AJR Am J Roentgenol,查看原文。版权归原作者所有。

引用

  1. Lee, Kim, Lee, et al. Large Language Model-Generated Expansion of the RadLex Ontology: Application to Multinational Datasets of Chest CT Reports. AJR Am J Roentgenol. 2026. PMID: 41532616. doi:10.2214/AJR.25.34243

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。