深度学习在超声甲状腺腺瘤样结节鉴别中的价值:双中心研究
本研究纳入648个甲状腺腺瘤样结节,分为训练、验证、外部测试和手术确认队列,比较深度学习迁移学习、影像组学及联合模型对良恶性鉴别的效能。结果显示深度学习迁移学习签名在测试队列AUC最高,为0.959,优于细针穿刺细胞学,特异度96.2%,总准确率89.3%。
要点
- 深度学习迁移学习签名测试队列AUC达0.959,高于影像组学和联合模型。
- 该签名特异度96.2%,总准确率89.3%,优于细针穿刺细胞学。
- 无需整合临床和影像组学特征,流程简化,适合临床决策辅助。
核心结论
- 深度学习迁移学习签名在甲状腺腺瘤样结节良恶性鉴别中性能稳健,测试队列AUC为0.959,优于细针穿刺细胞学。
研究设计与样本
- 双中心研究,纳入648个甲状腺腺瘤样结节,分为训练、验证、外部测试和手术确认队列。
- 使用PyRadiomics提取手工影像组学特征,采用预训练ResNet模型获取深度特征,经LASSO回归筛选稳健特征。
- 以AUC评估不同签名性能,并与细针穿刺细胞学诊断效能进行比较。
- 使用梯度加权类激活映射可视化深度学习的决策依据区域。
关键影像表现/征象
- 梯度加权类激活映射显示关键图像区域对签名决策有显著贡献。
诊断标准与关键数值
- 深度学习迁移学习签名在测试队列AUC为0.959。
- 深度学习影像组学签名测试队列AUC为0.937。
- 联合签名测试队列AUC为0.930。
- 影像组学签名测试队列AUC为0.835。
- 临床签名测试队列AUC为0.599。
临床应用与阅片要点
- 深度学习迁移学习签名无需整合临床和影像组学特征,流程简化,适合用于甲状腺腺瘤样结节良恶性鉴别的临床决策支持。
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
RATIONALE AND OBJECTIVES: Follicular neoplasms are difficult to classify by ultrasound, as they often present as thyroid adenomatoid nodules (TANU). Furthermore, fine-needle aspiration cytology (FNAC) exhibits limited accuracy in differentiating the nature of follicular lesions. This study aimed to develop and validate a novel model to optimize the workflow for distinguishing benign from malignant TANU.
MATERIALS AND METHODS: This study enrolled 648 TANUs, which were divided into the training (Train), validation (Val), external test (Test), and surgically confirmed (SC) cohorts. Handcrafted radiomics features were extracted using PyRadiomics, and deep features were obtained using pretrained ResNet models. Robust features were selected via LASSO regression. The deep learning radiomics (DLR) signature integrated the deep transfer learning (DTL) and radiomics signatures. The combined signature integrated the DLR and clinical signatures. The performance of different signatures was evaluated using the Area Under the Curve (AUC). We also compared the diagnostic performance of our best signature with FNAC. Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations were used to interpret the deep learning component.
RESULTS: The DTL signature achieved the highest AUC of 0.959 in the Test cohort (the DLR, combined, radiomics, and clinical signatures were 0.937, 0.930, 0.835, and 0.599, respectively). Furthermore, the DTL signature outperformed FNAC in diagnostic accuracy, with higher specificity (96.2%) and overall accuracy (89.3%). Grad-CAM confirmed that key image regions significantly contributed to the signature's decisions.
CONCLUSION: By eliminating the need to integrate clinical and radiomics features, the DTL signature streamlines the diagnostic process while maintaining robust performance, making it a highly effective tool for clinical decision-making.
引用
- Cheng, Liang, Zeng, et al. The Value of Deep Learning in Differentiating Thyroid Adenomatoid Nodules on Ultrasound: A Dual-Center Study. Acad Radiol. 2026. PMID: 42493350. doi:10.1016/j.acra.2026.06.047
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。