文献速递甲状腺与浅表技术与安全US

联合卷积神经网络融合超声图像与射频数据用于甲状腺结节识别

Thyroid nodule recognition using a joint convolutional neural network with information fusion of ultrasound images and radiofrequency data.
PubMed 文献 · Eur Radiol · 2021-01-06

研究纳入163对成人甲状腺结节的超声图像与射频信号,构建信息融合联合卷积神经网络(IF-JCNN)鉴别良恶性。五折交叉验证中,融合输入的准确率0.896、敏感度0.885、特异度0.910、AUROC 0.956,优于单用超声图像或单用射频信号。

要点

  • 融合超声图像与原始射频信号的IF-JCNN对甲状腺结节分类的AUROC达0.956。
  • 融合模型准确率0.896、敏感度0.885、特异度0.910,均优于单用超声图像或射频信号。
  • 超声成像前的原始射频信号可提供超声图像未携带的信息,两者具有互补性。

诊断标准与关键数值

  • IF-JCNN融合超声图像与射频信号:准确率0.896(95%CI 0.838-0.938)、敏感度0.885(95%CI 0.804-0.941)、特异度0.910(95%CI 0.815-0.966)、AUROC 0.956(95%CI 0.926-0.987)。
  • 仅用超声图像:准确率0.822(95%CI 0.755-0.878,p=0.0044)、敏感度0.792(95%CI 0.679-0.868,p=0.0091)、特异度0.866(95%CI 0.760-0.937,p=0.197)、AUROC 0.901(95%CI 0.855-0.948,p=0.0398)。
  • 仅用射频信号:准确率0.767(95%CI 0.694-0.829,p<0.001)、敏感度0.781(95%CI 0.685-0.859,p=0.0037)、特异度0.746(95%CI 0.625-0.845,p<0.001)、AUROC 0.845(95%CI 0.786-0.903,p<0.001)。

临床应用与阅片要点

  • 作者认为IF-JCNN模型填补了仅用超声图像卷积神经网络表征甲状腺结节的空白,可能作为辅助甲状腺癌诊断的工具。

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

OBJECTIVE: To develop a deep learning-based method with information fusion of US images and RF signals for better classification of thyroid nodules (TNs).

METHODS: One hundred sixty-three pairs of US images and RF signals of TNs from a cohort of adult patients were used for analysis. We developed an information fusion-based joint convolutional neural network (IF-JCNN) for the differential diagnosis of malignant and benign TNs. The IF-JCNN contains two branched CNNs for deep feature extraction: one for US images and the other one for RF signals. The extracted features are fused at the backend of IF-JCNN for TN classification.

RESULTS: Across 5-fold cross-validation, the accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) obtained by using the IF-JCNN with both US images and RF signals as inputs for TN classification were respectively 0.896 (95% CI 0.838-0.938), 0.885 (95% CI 0.804-0.941), 0.910 (95% CI 0.815-0.966), and 0.956 (95% CI 0.926-0.987), which were better than those obtained by using only US images: 0.822 (0.755-0.878; p = 0.0044), 0.792 (0.679-0.868, p = 0.0091), 0.866 (0.760-0.937, p = 0.197), and 0.901 (0.855-0.948, p = .0398), or RF signals: 0.767 (0.694-0.829, p < 0.001), 0.781 (0.685-0.859, p = 0.0037), 0.746 (0.625-0.845, p < 0.001), 0.845 (0.786-0.903, p < 0.001).

CONCLUSIONS: The proposed IF-JCNN model filled the gap of just using US images in CNNs to characterize TNs, and it may serve as a promising tool for assisting the diagnosis of thyroid cancer.

KEY POINTS: • Raw radiofrequency signals before ultrasound imaging of thyroid nodules provide useful information that is not carried by ultrasound images. • The information carried by raw radiofrequency signals and ultrasound images for thyroid nodules is complementary. • The performance of deep convolutional neural network for diagnosing thyroid nodules can be significantly improved by fusing US images and RF signals in the model as compared with just using US images.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Liu Z, Zhong S, Liu Q, et al. Thyroid nodule recognition using a joint convolutional neural network with information fusion of ultrasound images and radiofrequency data. Eur Radiol. 2021. PMID: 33409774. doi:10.1007/s00330-020-07585-z

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。