Deep Learning Based on ACR TI-RADS Can Improve the Differential Diagnosis of Thyroid Nodules

Frontiers in Oncology

Wen-Zhi Lv¹  Christoph F. Dietrich²  Bo Zhang³  Rui Yin⁴  Jian-Wei Xu⁵  Rui-Xue Chen⁶  Jia-Yu Wang⁷  Xin-Wu Cui⁷  Ge-Ge Wu⁷  Yu-Jing Yan⁷ 

[1] Department of Artificial Intelligence, Julei Technology Company, Wuhan, China;Department of General Internal Medicine, Kliniken Hirslanden Beau-Site,Bern, Switzerland;Department of Ultrasonic Imaging, Xiangya Hospital, Central South University, Changsha, China;Department of Ultrasound, Affiliated Renhe Hospital of China Three Gorges University, Yichang, China;Department of Ultrasound, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China;Department of Ultrasound, Wuchang Hospital, Wuhan University of Science and Technology, Wuhan, China;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;

关键词: artificial intelligence;    thyroid imaging reporting and data system (TI-RADS);    ultrasound;    thyroid cancer;    deep learning;   

DOI  :  10.3389/fonc.2021.575166

来源: DOAJ

【 摘 要 】

ObjectiveThe 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 MethodsFrom 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.ResultsIn 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 (P=0.108), 0.766 (P=0.101), 0.677 (P=0.211), 0.750 (P=0.128), and 0.680 (P=0.023), 0.761 (P=0.530), respectively.ConclusionsThe 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.

【 授权许可】

Unknown