基于TI-RADS词典的可解释超声影像组学甲状腺结节分类
本研究构建了将甲状腺超声定量影像组学特征映射到TI-RADS语义词典的可解释框架,联合三个多中心数据集共5542个结节。通过SHAP分析显示纹理异质性为主要恶性信号,Select-From-Model加Extra-Trees分类器在独立测试集上ROC-AUC达0.941±0.004。
要点
- 影像组学特征可通过TI-RADS词典直接对应超声语义,提升模型可解释性
- 纹理异质性、灰度游程矩阵非均匀性、强度离散度和峰度与高风险TI-RADS描述相关
- 模型在独立多中心测试集上达到ROC-AUC 0.941±0.004,提示稳定分类性能
核心结论
- 可解释影像组学词典结合稳定性感知模型选择框架,实现了基于超声的透明甲状腺结节风险分层
研究设计与样本
- 多中心回顾性研究,三个数据集合并后共5542个结节,70%用于五折交叉验证训练,30%作为独立多中心测试集
- 影像组学特征共107个,由PyRadiomics从二维超声图像提取,经min-max归一化
- 词典映射由四名医师、三名物理学家、一名放射学专家和一名生物学家通过专家共识完成
关键影像表现/征象
- SHAP分析显示纹理异质性为主导的恶性信号
- 灰度游程矩阵非均匀性、强度离散度和峰度与高风险TI-RADS描述符对齐
诊断标准与关键数值
- TI-RADS词典涵盖成分、回声、形态、边缘和回声灶等分类映射
- Select-From-Model(logistic regression)加Extra-Trees分类器测试集ROC-AUC为0.941±0.004
临床应用与阅片要点
- 影像组学输出可表达为TI-RADS术语,便于放射科医师理解模型决策依据
- 纹理异质性相关特征可作为甲状腺超声阅片时提示恶性的参考征象
局限性
- 原文未报告该词典在不同超声设备和操作者间的一致性验证
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
OBJECTIVE: Artificial Intelligence-based radiomics models for thyroid ultrasound (US) often lack interpretability, limiting clinical trust. This study develops and evaluates an interpretable radiomic feature (RF) framework for thyroid-nodule classification by linking quantitative-US features to the Thyroid-Imaging-Reporting-and-Data-System (TI-RADS) semantic lexicon through a clinically grounded radiomics dictionary.
METHODS: A radiomics dictionary was constructed to map TI-RADS categories, including composition, echogenicity, shape, margin, and echogenic foci, to Image-Biomarker-Standardization-Initiative-compliant RFs extracted from two-dimensional-US images. Relationships were defined through expert consensus (four physicians, three physicists, one radiology expert, one biologist) and further examined using Shapley-Additive-Explanations (SHAP) as a model-based interpretability analysis. Three multicenter datasets were combined, yielding 5,542 nodules, from which 107-RFs were extracted using PyRadiomics and normalized with min-max scaling. Twenty-seven feature-selection methods were paired with twenty-five classifiers and evaluated using stratified five-fold cross-validation on 70 % of the data, followed by evaluation on a held-out multicenter testing set comprising the remaining 30 % for benign-versus-malignant nodule classification. Robust model selection employed a stability-aware-composite-scoring framework combining mean performance and variability across accuracy, precision, recall, F1-score, and Receiver-Operating-Characteristic-Area-Under-the-Curve (ROC-AUC).
RESULTS: The dictionary enabled direct interpretation of radiomic signatures in TI-RADS terms. The Select-From-Model (logistic regression) plus Extra-Trees classifier achieved strong testing performance (ROC-AUC:0.941 ± 0.004). SHAP identified texture heterogeneity as the dominant malignancy signal, with Gray Level Run Length Matrix non-uniformity, intensity dispersion, and kurtosis aligning predictions with high-risk TI-RADS descriptors.
CONCLUSION: This study introduces an interpretable radiomics dictionary and stability-aware model selection framework, addressing interpretability limitations and enabling transparent thyroid nodule risk stratification from US.
引用
- Salmanpour, Taeb, Jouzdani, et al. A clinically anchored radiomics dictionary for explainable TI-RADS-based thyroid nodule classification in ultrasound; dictionary version TU1.0. Eur J Radiol. 2026. PMID: 42302438. doi:10.1016/j.ejrad.2026.113014
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。