深度学习语言建模方法用于自动化、个性化及迭代式放射-病理对照
本研究采用通用语言模型微调方法,用20万份放射与病理报告适应放射-病理空间,并以10万对匹配/不匹配候选对训练二分类模型。模型与上一代 anatomic concept 匹配系统高度一致,特异度100%、敏感度65.1%、准确率73.7%,提示深度学习语言建模用于放射-病理对照有前景。
要点
- 深度学习语言模型可快速适应既有标注,与上一代解剖概念匹配系统表现高度接近。
- 本模型特异度100%、敏感度65.1%、准确率73.7%,对照方法分别为98.0%、65.1%、73.2%。
- 模型可连续改进和调优,以适应使用者和站点层面的偏好。
核心结论
- 现代深度学习语言建模方法对放射-病理对照有前景,可快速适应底层训练标签并持续改进与调优。
研究设计与样本
- 采用通用语言模型微调(Universal Language Model Fine-Tuning for Text Classification)方法,用20万份放射与病理报告适应放射-病理空间,以10万对匹配与不匹配各半的候选对训练二分类模型,匹配由上一代人工智能解剖概念放射-病理对照系统定义。
诊断标准与关键数值
- 本模型以既往解剖概念匹配系统定义为参考,特异度100%、敏感度65.1%、准确率73.7%;对照方法特异度98.0%、敏感度65.1%、准确率73.2%。
临床应用与阅片要点
- 作者认为现代深度学习语言建模方法对放射-病理对照有前景,模型可连续改进和调优以提升性能并适应使用者和站点层面偏好。
局限性
- 匹配定义依赖上一代人工智能解剖概念放射-病理对照系统。
- 原文未报告其他局限性。
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
PURPOSE: Radiology-pathology correlation has long been foundational to continuing education, peer learning, quality assurance, and multidisciplinary patient care. The objective of this study was to determine whether modern deep-learning language-modeling techniques could reliably match pathology reports to pertinent radiology reports.
METHODS: The recently proposed Universal Language Model Fine-Tuning for Text Classification methodology was used. Two hundred thousand radiology and pathology reports were used for adaptation to the radiology-pathology space. One hundred thousand candidate radiology-pathology pairs, evenly split into match and no-match categories, were used for training the final binary classification model. Matches were defined by a previous-generation artificial intelligence anatomic concept radiology-pathology correlation system.
RESULTS: The language model rapidly adapted very closely to the prior anatomic concept-matching approach, with 100% specificity, 65.1% sensitivity, and 73.7% accuracy. For comparison, the previous methodology, which was intentionally designed to be specific at the expense of sensitivity, had 98.0% specificity, 65.1% sensitivity, and 73.2% accuracy.
CONCLUSIONS: Modern deep-learning language-modeling approaches are promising for radiology-pathology correlation. Because of their rapid adaptation to underlying training labels, these models advance previous artificial intelligence work in that they can be continuously improved and tuned to improve performance and adjust to user and site-level preference.
引用
- Filice RW. Deep-Learning Language-Modeling Approach for Automated, Personalized, and Iterative Radiology-Pathology Correlation. J Am Coll Radiol. 2019. PMID: 31173746. doi:10.1016/j.jacr.2019.05.007
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。