Ultrasound-based deep learning in the establishment of a breast lesion risk stratification system: a multicenter study.

Autor: Gu Y; Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Shuai Fu Yuan, Dong Cheng District, Beijing, 100730, China., Xu W; Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Shuai Fu Yuan, Dong Cheng District, Beijing, 100730, China., Liu T; Department of Medical Imaging Advanced Research, Beijing Research Institute, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Beijing, China., An X; Department of Medical Imaging Advanced Research, Beijing Research Institute, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Beijing, China., Tian J; Department of Ultrasound, The Second Affiliated Hospital of Harbin Medical University, Harbin, China., Ran H; Department of Ultrasound, The Second Affiliated Hospital of Chongqing Medical University & Chongqing Key Laboratory of Ultrasound Molecular Imaging, Chongqing, China., Ren W; Department of Ultrasound, Shengjing Hospital of China Medical University, Shenyang, China., Chang C; Department of Medical Ultrasound, Fudan University Shanghai Cancer Center, Fudan University, Shanghai, China., Yuan J; Department of Ultrasonography, Henan Provincial People's Hospital, Zhengzhou, China., Kang C; Department of Ultrasound, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Taiyuan, China., Deng Y; Department of Medical Ultrasound, Tongji Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, China., Wang H; Department of Ultrasound, China-Japan Union Hospital of Jilin University, Changchun, China., Luo B; Department of Ultrasound, The Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China., Guo S; Department of Ultrasonography, First Affiliated Hospital of Guangxi Medical University, Nanning, China., Zhou Q; Department of Medical Ultrasound, The Second Affiliated Hospital, School of Medicine, Xi'an Jiaotong University, Xi'an, China., Xue E; Department of Ultrasound, Union Hospital of Fujian Medical University, Fujian Institute of Ultrasound Medicine, Fuzhou, China., Zhan W; Department of Ultrasound, Ruijin Hospital, Shanghai Jiaotong University, School of Medicine, Shanghai, China., Zhou Q; Department of Ultrasonography, Renmin Hospital of Wuhan University, Wuhan, China., Li J; Department of Ultrasound, Qilu Hospital, Shandong University, Jinan, China., Zhou P; Department of Ultrasound, The Third Xiangya Hospital of Central South University, Changsha, China., Chen M; Department of Ultrasound Medicine, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China., Gu Y; Department of Ultrasonography, The Affiliated Hospital of Guizhou Medical University, Guiyang, China., Chen W; Department of Ultrasound, The First Hospital of Shanxi Medical University, Taiyuan, China., Zhang Y; Department of Ultrasound, The Second Hospital of Dalian Medical University, Dalian, China., Li J; Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Shuai Fu Yuan, Dong Cheng District, Beijing, 100730, China., Cong L; Department of Medical Imaging Advanced Research, Beijing Research Institute, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Beijing, China., Zhu L; Department of Medical Imaging Advanced Research, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China., Wang H; Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Shuai Fu Yuan, Dong Cheng District, Beijing, 100730, China. whychina@126.com., Jiang Y; Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Shuai Fu Yuan, Dong Cheng District, Beijing, 100730, China. jiangyuxinxh@163.com.
Jazyk: angličtina
Zdroj: European radiology [Eur Radiol] 2023 Apr; Vol. 33 (4), pp. 2954-2964. Date of Electronic Publication: 2022 Nov 23.
DOI: 10.1007/s00330-022-09263-8
Abstrakt: Objectives: To establish a breast lesion risk stratification system using ultrasound images to predict breast malignancy and assess Breast Imaging Reporting and Data System (BI-RADS) categories simultaneously.
Methods: This multicenter study prospectively collected a dataset of ultrasound images for 5012 patients at thirty-two hospitals from December 2018 to December 2020. A deep learning (DL) model was developed to conduct binary categorization (benign and malignant) and BI-RADS categories (2, 3, 4a, 4b, 4c, and 5) simultaneously. The training set of 4212 patients and the internal test set of 416 patients were from thirty hospitals. The remaining two hospitals with 384 patients were used as an external test set. Three experienced radiologists performed a reader study on 324 patients randomly selected from the test sets. We compared the performance of the DL model with that of three radiologists and the consensus of the three radiologists.
Results: In the external test set, the DL model achieved areas under the receiver operating characteristic curve (AUCs) of 0.980 and 0.945 for the binary categorization and six-way categorizations, respectively. In the reader study set, the DL BI-RADS categories achieved a similar AUC (0.901 vs. 0.933, p = 0.0632), sensitivity (90.98% vs. 95.90%, p = 0.1094), and accuracy (83.33% vs. 79.01%, p = 0.0541), but higher specificity (78.71% vs. 68.81%, p = 0.0012) than those of the consensus of the three radiologists.
Conclusions: The DL model performed well in distinguishing benign from malignant breast lesions and yielded outcomes similar to experienced radiologists. This indicates the potential applicability of the DL model in clinical diagnosis.
Key Points: • The DL model can achieve binary categorization for benign and malignant breast lesions and six-way BI-RADS categorizations for categories 2, 3, 4a, 4b, 4c, and 5, simultaneously. • The DL model showed acceptable agreement with radiologists for the classification of breast lesions. • The DL model performed well in distinguishing benign from malignant breast lesions and had promise in helping reduce unnecessary biopsies of BI-RADS 4a lesions.
(© 2022. The Author(s), under exclusive licence to European Society of Radiology.)
Databáze: MEDLINE
Externí odkaz: