HwaMei Hospital, University of Chinese Academy of Sciences, 41 Xibei Street, Ningbo, 315010, China; CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences, 95 Zhongguancun East Road, Beijing, 100190, China; School of Artificial Intelligence, University of Chinese Academy of Sciences, No.19 (A) Yuquan Road, Shijingshan District, Beijing, 100049, China.
HwaMei Hospital, University of Chinese Academy of Sciences, 41 Xibei Street, Ningbo, 315010, China.
Eur J Radiol. 2020 Jun;127:108992. doi: 10.1016/j.ejrad.2020.108992. Epub 2020 Apr 12.
We aimed to propose a highly automatic and objective model named deep learning Radiomics of thyroid (DLRT) for the differential diagnosis of benign and malignant thyroid nodules from ultrasound (US) images.
We retrospectively enrolled and finally include US images and fine-needle aspiration biopsies from 1734 patients with 1750 thyroid nodules. A basic convolutional neural network (CNN) model, a transfer learning (TL) model, and a newly designed model named deep learning Radiomics of thyroid (DLRT) were used for the investigation. Their diagnostic accuracy was further compared with human observers (one senior and one junior US radiologist). Moreover, the robustness of DLRT over different US instruments was also validated. Analysis of receiver operating characteristic (ROC) curves were performed to calculate optimal area under it (AUC) for benign and malignant nodules. One observer helped to delineate the nodules.
AUCs of DLRT were 0.96 (95% confidence interval [CI]: 0.94-0.98), 0.95 (95% confidence interval [CI]: 0.93-0.97) and 0.97 (95% confidence interval [CI]: 0.95-0.99) in the training, internal and external validation cohort, respectively, which were significantly better than other deep learning models (P < 0.01) and human observers (P < 0.001). No significant difference was found when applying DLRT on thyroid US images acquired from different US instruments.
DLRT shows the best overall performance comparing with other deep learning models and human observers. It holds great promise for improving the differential diagnosis of benign and malignant thyroid nodules.
我们旨在提出一种名为深度学习甲状腺放射组学（DLRT）的高度自动化和客观模型，用于从超声（US）图像中鉴别诊断良性和恶性甲状腺结节。
我们回顾性地招募了最终包括 1734 名患者的 1750 个甲状腺结节的 US 图像和细针抽吸活检。使用基本卷积神经网络（CNN）模型、迁移学习（TL）模型和新设计的模型深度学习甲状腺放射组学（DLRT）进行研究。进一步将其诊断准确性与人类观察者（一名高级和一名初级 US 放射科医生）进行比较。此外，还验证了 DLRT 在不同 US 仪器上的稳健性。进行接收器操作特征（ROC）曲线分析以计算良性和恶性结节的最佳曲线下面积（AUC）。一名观察者帮助描绘结节。
DLRT 在训练、内部和外部验证队列中的 AUC 分别为 0.96（95%置信区间[CI]：0.94-0.98）、0.95（95%置信区间[CI]：0.93-0.97）和 0.97（95%置信区间[CI]：0.95-0.99），均显著优于其他深度学习模型（P < 0.01）和人类观察者（P < 0.001）。在从不同 US 仪器获取的甲状腺 US 图像上应用 DLRT 时，未发现显著差异。
与其他深度学习模型和人类观察者相比，DLRT 显示出最佳的整体性能。它有望改善良性和恶性甲状腺结节的鉴别诊断。
