文献速递甲状腺与浅表US

机器学习辅助多普勒特征提升甲状腺癌诊断:多队列研究

Machine Learning Assisted Doppler Features for Enhancing Thyroid Cancer Diagnosis: A Multi-Cohort Study.
PubMed 文献 · J Ultrasound Med · 2021-11-09

本回顾性多队列研究纳入674例患者共712个甲状腺结节,利用人工神经网络结合彩色多普勒超声定量特征与ACR TI-RADS灰度特征构建TDUS-Net模型。结果显示TDUS-Net在内部测试中AUC为0.898,高于仅用灰度特征的TUS-Net(0.881);外部测试中TDUS-Net的AUC为0.925,优于放射科医师的0.810。

要点

  • 联合彩色多普勒定量特征与灰度超声特征可提升甲状腺结节分类性能。
  • TDUS-Net在外部测试中AUC达0.925,高于放射科医师的0.810。
  • 多普勒特征包括整结节、结节内、周边血流比例及血管数量。

核心结论

  • 结合灰度超声特征与彩色多普勒超声特征的人工神经网络模型在甲状腺结节良恶性分类中可达到与放射科医师相当或更高的性能。

研究设计与样本

  • 回顾性多队列研究,纳入674例患者共712个甲状腺结节,其中内部数据集512个、外部数据集200个。
  • 使用人工神经网络构建TDUS-Net模型,输入包括自动提取的定量彩色多普勒超声特征(整结节血流比例、结节内血流比例、周边血流比例、血管数量)和ACR TI-RADS定义的灰度超声特征。
  • 对照模型为仅基于灰度超声特征的TUS-Net,并比较了放射科医师的诊断性能。

诊断标准与关键数值

  • 内部测试中TDUS-Net的AUC为0.898(95%CI:0.868-0.922),TUS-Net的AUC为0.881(95%CI:0.850-0.908)。
  • 外部测试中TDUS-Net的AUC为0.925(95%CI:0.880-0.958),放射科医师的AUC为0.810(95%CI:0.749-0.862)。
  • 灰度超声特征依据ACR TI-RADS标准定义。

临床应用与阅片要点

  • 在甲状腺结节超声评估中,可关注彩色多普勒定量特征(如血流比例和血管数量)作为灰度特征的补充。
  • 机器学习模型可作为辅助工具,帮助提升甲状腺癌分类的诊断性能。

局限性

  • 本研究为回顾性设计,样本来自随机选取的患者,可能存在选择偏倚。
  • 样本量有限(共712个结节),外部验证集仅200个结节。

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: This pilot study aims at exploiting machine learning techniques to extract color Doppler ultrasound (CDUS) features and to build an artificial neural network (ANN) model based on these CDUS features for improving the diagnostic performance of thyroid cancer classification.

METHODS: A total of 674 patients with 712 thyroid nodules (TNs) (512 from internal dataset and 200 from external dataset) were randomly selected in this retrospective study. We used ANN to build a model (TDUS-Net) for classifying malignant and benign TNs using both the automatically extracted quantitative CDUS features (whole ratio, intranodular ratio, peripheral ratio, and number of vessels) and gray-scale ultrasound (US) features defined by the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Then, we compared the diagnostic performance of the model, the performance of another ANN model based on the gray-scale US features alone (TUS-Net), and that of radiologists.

RESULTS: The TDUS-Net (0.898, 95% CI: 0.868-0.922) achieved a higher area under the curve (AUC) than that of TUS-Net (0.881, 95% CI: 0.850-0.908) in the internal tests. Compared with radiologists, TDUS-Net (AUC: 0.925, 95% CI: 0.880-0.958) performed better than radiologists (AUC: 0.810, 95% CI: 0.749-0.862) in the external tests.

CONCLUSIONS: Applying a machine learning model by combining both gray-scale US features and CDUS features can achieve comparable or even higher performance than radiologists in classifying TNs.

来源:J Ultrasound Med,查看原文。版权归原作者所有。

引用

  1. Zhu, Du, Jiang, et al. Machine Learning Assisted Doppler Features for Enhancing Thyroid Cancer Diagnosis: A Multi-Cohort Study. J Ultrasound Med. 2021. PMID: 34751458. doi:10.1002/jum.15873

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。