基于美国放射学会甲状腺影像报告和数据系统的深度学习可改善甲状腺结节的鉴别诊断。
Deep Learning Based on ACR TI-RADS Can Improve the Differential Diagnosis of Thyroid Nodules.
作者信息
Wu Ge-Ge, Lv Wen-Zhi, Yin Rui, Xu Jian-Wei, Yan Yu-Jing, Chen Rui-Xue, Wang Jia-Yu, Zhang Bo, Cui Xin-Wu, Dietrich Christoph F
机构信息
Sino-German Tongji-Caritas Research Center of Ultrasound in Medicine, Department of Medical Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Department of Artificial Intelligence, Julei Technology Company, Wuhan, China.
出版信息
Front Oncol. 2021 Apr 27;11:575166. doi: 10.3389/fonc.2021.575166. eCollection 2021.
OBJECTIVE
The purpose of this study was to improve the differentiation between malignant and benign thyroid nodules using deep learning (DL) in category 4 and 5 based on the Thyroid Imaging Reporting and Data System (TI-RADS, TR) from the American College of Radiology (ACR).
DESIGN AND METHODS
From June 2, 2017 to April 23, 2019, 2082 thyroid ultrasound images from 1396 consecutive patients with confirmed pathology were retrospectively collected, of which 1289 nodules were category 4 (TR4) and 793 nodules were category 5 (TR5). Ninety percent of the B-mode ultrasound images were applied for training and validation, and the residual 10% and an independent external dataset for testing purpose by three different deep learning algorithms.
RESULTS
In the independent test set, the DL algorithm of best performance got an AUC of 0.904, 0.845, 0.829 in TR4, TR5, and TR4&5, respectively. The sensitivity and specificity of the optimal model was 0.829, 0.831 on TR4, 0.846, 0.778 on TR5, 0.790, 0.779 on TR4&5, versus the radiologists of 0.686 (=0.108), 0.766 (=0.101), 0.677 (=0.211), 0.750 (=0.128), and 0.680 (=0.023), 0.761 (=0.530), respectively.
CONCLUSIONS
The study demonstrated that DL could improve the differentiation of malignant from benign thyroid nodules and had significant potential for clinical application on TR4 and TR5.
目的
本研究旨在基于美国放射学会（ACR）的甲状腺影像报告和数据系统（TI-RADS，TR），利用深度学习（DL）提高4类和5类甲状腺结节良恶性的鉴别能力。
设计与方法
回顾性收集2017年6月2日至2019年4月23日期间1396例经病理确诊的连续患者的2082幅甲状腺超声图像，其中1289个结节为4类（TR4），793个结节为5类（TR5）。90%的B超图像用于训练和验证，其余10%以及一个独立的外部数据集用于三种不同深度学习算法的测试。
结果
在独立测试集中，性能最佳的DL算法在TR4、TR5和TR4&5中的曲线下面积（AUC）分别为0.904、0.845和0.829。最佳模型在TR4上的敏感性和特异性分别为0.829、和0.831，在TR5上为0.846、0.778，在TR4&5上为0.790、0.779，而放射科医生的相应数据分别为0.686（±0.108）、0.766（±0.101），0.677（±0.211）、0.750（±0.128），以及0.680（±0.023）、0.761（±0.530）。
结论
该研究表明，DL可以提高甲状腺结节良恶性的鉴别能力，在TR4和TR5的临床应用中具有显著潜力。
Zotero 插件