基于知识图谱的小样本学习用于医学影像报告标注
本研究构建腰椎X线影像知识图谱,标注2000条数据并按7:2:1划分为训练、验证与测试集,利用知识图谱的同义词与替换属性扩增训练数据并输入BERT模型进行自动标注训练;扩增后精确率由0.728升至0.932(扩增比1:100),F1由0.666升至0.881,补充验证集表现不佳节点的数据后平均F1达0.979。
要点
- 知识图谱同义词与替换联合扩增(1:100)F1=0.881,优于仅同义词(0.815)与仅替换(0.753)。
- 扩增比1:100较1:10精确率0.932对0.912、F1 0.881对0.853,差异有统计学意义(P<.05)。
- 补充验证集表现不佳节点的数据后平均F1升至0.979(P<.05)。
核心结论
- 基于知识图谱以小样本数据集训练影像报告自动标注模型,可减少人工标注工作量并提高影像数据标注的效率与准确性。
研究设计与样本
- 研究构建腰椎X线影像知识图谱,并基于该图谱标注2000条数据,按7:2:1分为训练集、验证集和测试集。
- 训练集基于知识图谱的同义词/替换属性进行扩增,扩增数据输入BERT模型进行自动标注训练。
- 采用精确率与F1评分评估不同扩增比(1:10、1:100、1:1000)及不同扩增方法(仅同义词、仅替换、同义词与替换联合)下的模型性能。
诊断标准与关键数值
- 扩增前精确率0.728、F1评分0.666。
- 扩增比1:10时精确率0.912、F1评分0.853;扩增比1:100时精确率0.932、F1评分0.881(P<.05)。
- 扩增比1:100时,同义词与替换联合法F1=0.881,优于仅同义词法F1=0.815与仅替换法F1=0.753(P<.05)。
- 补充验证集表现不佳节点的数据后,平均F1评分提高至0.979(P<.05)。
临床应用与阅片要点
- 作者认为该方法可减少人工标注工作量,提高影像数据标注的效率与准确性,为AI在影像报告自动标注领域的应用提供研究策略。
局限性
- 原文未报告局限性。
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
BACKGROUND: The application of artificial intelligence (AI) in the field of automatic imaging report labeling faces the challenge of manually labeling large datasets.
PURPOSE: To propose a data augmentation method by using knowledge graph (KG) and few-shot learning.
METHODS: A KG of lumbar spine X-ray images was constructed, and 2000 data were annotated based on the KG, which were divided into training, validation, and test sets in a ratio of 7:2:1. The training dataset was augmented based on the synonym/replacement attributes of the KG and was the augmented data was input into the BERT (Bidirectional Encoder Representations from Transformers) model for automatic annotation training. The performance of the model under different augmentation ratios (1:10, 1:100, 1:1000) and augmentation methods (synonyms only, replacements only, combination of synonyms and replacements) was evaluated using the precision and F1 scores. In addition, with the augmentation ratio was fixed, iterative experiments were performed by supplementing the data of nodes that perform poorly in the validation set to further improve model's performance.
RESULTS: Prior to data augmentation, the precision was 0.728 and the F1 score was 0.666. By adjusting the augmentation ratio, the precision increased from 0.912 at a 1:10 augmentation ratio to 0.932 at a 1:100 augmentation ratio (P<.05), while F1 score improved from 0.853 at a 1:10 augmentation ratio to 0.881 at a 1:100 augmentation ratio (P<.05). Additionally, the effectiveness of various augmentation methods was compared at a 1:100 augmentation ratio. The augmentation method that combined synonyms and replacements (F1=0.881) was superior to the methods that only used synonyms (F1=0.815) and only used replacements (F1=0.753) (P<.05). For nodes that exhibited suboptimal performance on the validation set, supplementing the training set with target data improved model performance, increasing the average F1 score to 0.979 (P<.05).
CONCLUSION: Based on the KG, this study trained an automatic labeling model of radiology reports using a few-shot data set. This method effectively reduces the workload of manual labeling, improves the efficiency and accuracy of image data labeling, and provides an important research strategy for the application of AI in the domain of automatic labeling of image reports.
引用
- Li T, Zhang Y, Su D, et al. Knowledge Graph-Based Few-Shot Learning for Label of Medical Imaging Reports. Acad Radiol. 2025. PMID: 40140273. doi:10.1016/j.acra.2025.02.045
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。