Research paper Improving B-mode ultrasound diagnostic performance for focal liver lesions using deep learning: A multicentre study

doi:10.1016/j.ebiom.2020.102777

CORC > 自动化研究所 > 中国科学院自动化研究所 > 中国科学院分子影像重点实验室

	Research paper Improving B-mode ultrasound diagnostic performance for focal liver lesions using deep learning: A multicentre study
	Yang, Qi 2; Wei, Jingwei 3,4; Hao, Xiaohan 3,4,5; Kong, Dexing 6; Yu, Xiaoling 2; Jiang, Tianan 7; Xi, Junqing 2; Cai, Wenjia 2; Luo, Yanchun 2; Jing, Xiang 8
刊名	EBIOMEDICINE
	2020-06-01
卷号	56 页码:9
关键词	Ultrasound Convolutional neural network Focal liver lesions Diagnosis
ISSN号	2352-3964
DOI	10.1016/j.ebiom.2020.102777
通讯作者	Zheng, Rongqin(zhengrq@mail.sysu.edu.cn) ; Yu, Jie(jiemi301@163.com) ; Tian, Jie(jie.tian@ia.ac.cn) ; Liang, Ping(liangping301@hotmail.com)
英文摘要	Background: The diagnosis performance of B-mode ultrasound (US) for focal liver lesions (FLLs) is relatively limited. We aimed to develop a deep convolutional neural network of US (DCNN-US) for aiding radiologists in classification of malignant from benign FLLs. Materials and methods: This study was conducted in 13 hospitals and finally 2143 patients with 24,343 US images were enrolled. Patients who had non-cystic FLLs with pathological results were enrolled. The FLLs from 11 hospitals were randomly divided into training and internal validations (IV) cohorts with a 4:1 ratio for developing and evaluating DCNN-US. Diagnostic performance of the model was verified using external validation (EV) cohort from another two hospitals. The diagnosis value of DCNN-US was compared with that of contrast enhanced computed tomography (CT)/magnetic resonance image (MRI) and 236 radiologists, respectively. Findings: The AUC of Model(LBC) for FLLs was 0.924 (95% CI: 0.889-0.959) in the EV cohort. The diagnostic sensitivity and specificity of ModelLBC were superior to 15-year skilled radiologists (86.5% vs 76.1%, p = 0.0084 and 85.5% vs 76.9%, p = 0.0051, respectively). Accuracy of ModelLBC was comparable to that of contrast enhanced CT (both 84.7%) but inferior to contrast enhanced MRI (87.9%) for lesions detected by US. Interpretation: DCNN-US with high sensitivity and specificity in diagnosing FLLs shows its potential to assist less-experienced radiologists in improving their performance and lowering their dependence on sectional imaging in liver cancer diagnosis. (C) 2020 The Authors. Published by Elsevier B.V.
资助项目	National Scientific Foundation Committee of China[81627803] ; National Scientific Foundation Committee of China[81971625] ; National Scientific Foundation Committee of China[91859201] ; National Scientific Foundation Committee of China[81227901] ; National Scientific Foundation Committee of China[81527805] ; National Scientific Foundation Committee of Beijing[JQ18021] ; Fostering Funds for National Distinguished Young Scholar Science Fund[NCRCG-PLAGH-2019011] ; National Clinical Research Centre for Geriatric Diseases of Chinese PLA General Hospital ; National Key R&D Program of Ministry of Science and Technology of China[2018ZX10723-204] ; Chinese Academy of Sciences[GJJSTD20170004] ; Chinese Academy of Sciences[QYZDJ-SSW-JSC005] ; Beijing Municipal Science & Technology Commission[Z171100000117023] ; Strategic Priority Research Program of Chinese Academy of Science[XDBS01000000]
WOS研究方向	General & Internal Medicine ; Research & Experimental Medicine
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000549929200011
资助机构	National Scientific Foundation Committee of China ; National Scientific Foundation Committee of Beijing ; Fostering Funds for National Distinguished Young Scholar Science Fund ; National Clinical Research Centre for Geriatric Diseases of Chinese PLA General Hospital ; National Key R&D Program of Ministry of Science and Technology of China ; Chinese Academy of Sciences ; Beijing Municipal Science & Technology Commission ; Strategic Priority Research Program of Chinese Academy of Science
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/40161]
专题	自动化研究所_中国科学院分子影像重点实验室
通讯作者	Zheng, Rongqin; Yu, Jie; Tian, Jie; Liang, Ping
作者单位	1.Fujian Canc Hosp, Dept Ultrasound, Fuzhou, Peoples R China 2.Chinese Peoples Liberat Army Gen Hosp, Dept Intervent Ultrasound, 28 Fuxing Rd, Beijing 100853, Peoples R China 3.Chinese Acad Sci, Key Lab Mol Imaging, Inst Automat, Beijing, Peoples R China 4.Univ Chinese Acad Sci, Beijing, Peoples R China 5.Univ Sci & Technol China, Ctr Biomed Engn, Hefei, Peoples R China 6.Zhejiang Univ, Sch Math Sci, Hangzhou, Peoples R China 7.Zhejiang Univ, Coll Med, Affiliated Hosp 1, Dept Ultrasound, Hangzhou, Jiangsu, Peoples R China 8.Tianjin Third Cent Hosp, Dept Ultrasound, Tianjin, Peoples R China 9.Fourth Mil Med Univ, Tangdu Hosp, Dept Ultrasound Diag, Xian, Peoples R China 10.Harbin First Hosp, Dept Ultrasound, Harbin, Peoples R China
推荐引用方式 GB/T 7714	Yang, Qi,Wei, Jingwei,Hao, Xiaohan,et al. Research paper Improving B-mode ultrasound diagnostic performance for focal liver lesions using deep learning: A multicentre study[J]. EBIOMEDICINE,2020,56:9.
APA	Yang, Qi.,Wei, Jingwei.,Hao, Xiaohan.,Kong, Dexing.,Yu, Xiaoling.,...&Liang, Ping.(2020).Research paper Improving B-mode ultrasound diagnostic performance for focal liver lesions using deep learning: A multicentre study.EBIOMEDICINE,56,9.
MLA	Yang, Qi,et al."Research paper Improving B-mode ultrasound diagnostic performance for focal liver lesions using deep learning: A multicentre study".EBIOMEDICINE 56(2020):9.