[论文解读] A Clinically Anchored Radiomics Dictionary for Explainable TI-RADS-Based Thyroid Nodule Classification in Ultrasound; Dictionary Version TU1.0

Mohammad Salmanpour, Shahram Taeb+6 more|arXiv (Cornell University)|Mar 24, 2026

Thyroid Cancer Diagnosis and Treatment被引用 0

一句话总结

该论文提出一个 clinically anchored 的放射组学框架，将超声放射组学特征与 TI-RADS 语义联系起来，在多中心数据上实现可解释的甲状腺结节分类，达到高 ROC-AUC。

ABSTRACT

Artificial intelligence based radiomics models for thyroid ultrasound (US) often achieve strong diagnostic performance but remain difficult to interpret, limiting clinical trust and adoption. We developed and validated 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. The dictionary mapped 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 and examined using Shapley Additive Explanations (SHAP). Three multicenter datasets were combined, yielding 5,542 nodules. A total of 107 RFs were extracted using PyRadiomics and normalized with min-max scaling. For benign versus malignant classification, 27 feature selection methods were paired with 25 classifiers and evaluated using stratified five-fold cross-validation on 70% of the data, followed by testing on the remaining 30%. Robust model selection used a stability-aware composite score combining mean performance and variability across balanced accuracy, precision, recall, F1-score, and ROC-AUC. The proposed dictionary enabled direct interpretation of radiomic signatures in TI-RADS terms. The best model, Select-From-Model based on logistic regression with Extra-Trees, achieved a test ROC-AUC of 0.941 +/- 0.005. SHAP analysis showed that texture heterogeneity was the dominant malignancy signal, with gray level run length matrix non-uniformity, intensity dispersion, and kurtosis aligning with high-risk TI-RADS descriptors. These findings support transparent and clinically meaningful thyroid nodule risk stratification from US.

研究动机与目标

• 将定量放射组学特征与 TI-RADS 语义描述符结合，以提升甲状腺超声分类的可解释性和临床信任度。
• 开发并验证将 TI-RADS 分类映射到 Image Biomarker Standardization Initiative (IBSI) 兼容放射组学特征的词典。
• 在多中心数据上评估跨多种特征选择方法和分类器的稳健模型选择。
• 通过 SHAP 分析将放射组学特征与 TI-RADS 风险描述符联系起来，展示可解释性。

提出的方法

• 使用 PyRadiomics 从二维超声图像提取 107 个放射组学特征，并进行最小-最大归一化。
• 构建将 TI-RADS 分类（组成、回声性、形状、边缘、回声灶）映射到 RF 的词典。
• 将三个多中心数据集合并，总计 5,542 例用于训练与测试。
• 在分层五折交叉验证（70% 训练、30% 测试）下，评估 27 种特征选择方法和 25 种分类器。
• 使用稳定性感知的综合评分在平衡准确率、精准率、召回率、F1、ROC-AUC 之间选取稳健模型。
• 应用 SHAP 来解释特征贡献并将其与 TI-RADS 描述符关联起来。

实验结果

研究问题

• RQ1 基于 TI-RADS 语义的放射组学词典能否在超声中实现可解释的甲状腺结节分类？
• RQ2 根据 SHAP，哪些放射组学特征与 TI-RADS 描述符最强烈指示恶性？
• RQ3 在多中心数据上，基于字典引导的稳健模型在良性 vs 恶性甲状腺结节上的性能如何？

主要发现

• 最佳模型（From-Model 选择与逻辑回归和额外树集成）在测试集上的 ROC-AUC 为 0.941 ± 0.005。
• 共提取并在 27 种特征选择器和 25 种分类器中使用了 107 个放射组学特征。
• SHAP 将纹理异质性识别为主导的恶性信号；如灰度等级运行长度矩阵的非均匀性、强度离散度、峰度等特征与高风险 TI-RADS 描述符一致。
• 该词典使放射组学特征能直接用 TI-RADS 术语解释，支持透明的甲状腺结节风险分层。