A deep learning radiomics model for preoperative grading in meningioma

doi:10.1016/j.ejrad.2019.04.022

	A deep learning radiomics model for preoperative grading in meningioma
	Zhu, Yongbei 1,2,3; Man, Chuntao 1; Gong, Lixin 4; Dong, Di2,5 ; Yu, Xinyi 6; Wang, Shuo2,5 ; Fang, Mengjie2,5 ; Wang, Siwen 2,5; Fang, Xiangming 6; Chen, Xuzhu 7
刊名	EUROPEAN JOURNAL OF RADIOLOGY
	2019-07-01
卷号	116 页码:128-134
关键词	Radiomics Deep learning Meningioma Tumor grading Magnetic resonance imaging
ISSN号	0720-048X
DOI	10.1016/j.ejrad.2019.04.022
通讯作者	Fang, Xiangming(drfxm@163.com) ; Chen, Xuzhu(radiology888@aliyun.com) ; Tian, Jie(tian@ieee.org)
英文摘要	Objectives: To noninvasively differentiate meningioma grades by deep learning radiomics (DLR) model based on routine post-contrast MRI. Methods: We enrolled 181 patients with histopathologic diagnosis of meningioma who received post-contrast MRI preoperative examinations from 2 hospitals (99 in the primary cohort and 82 in the validation cohort). All the tumors were segmented based on post-contrast axial T1 weighted images (T1WI), from which 2048 deep learning features were extracted by the convolutional neural network. The random forest algorithm was used to select features with importance values over 0.001, upon which a deep learning signature was built by a linear discriminant analysis classifier. The performance of our DLR model was assessed by discrimination and calibration in the independent validation cohort. For comparison, a radiomic model based on hand-crafted features and a fusion model were built. Results: The DLR signature comprised 39 deep learning features and showed good discrimination performance in both the primary and validation cohorts. The area under curve (AUC), sensitivity, and specificity for predicting meningioma grades were 0.811(95% CI, 0.635-0.986), 0.769, and 0.898 respectively in the validation cohort. DLR performance was superior over the hand-crafted features. Calibration curves of DLR model showed good agreements between the prediction probability and the observed outcome of high-grade meningioma. Conclusions: Using routine MRI data, we developed a DLR model with good performance for noninvasively individualized prediction of meningioma grades, which achieved a quantization capability superior over the hand-crafted features. This model has potential to guide and facilitate the clinical decision-making of whether to observe or to treat patients by providing prognostic information.
资助项目	National Key R&D Program of China[2017YFC1308700] ; National Key R&D Program of China[2017YFA0205200] ; National Key R&D Program of China[2017YFC1309100] ; National Key R&D Program of China[2017YFC0114300] ; National Key R&D Program of China[2018YFC0115604] ; National Natural Science Foundation of China[81771924] ; National Natural Science Foundation of China[81501616] ; National Natural Science Foundation of China[81227901] ; National Natural Science Foundation of China[81671854] ; National Natural Science Foundation of China[81772005] ; National Natural Science Foundation of China[81271629] ; Beijing Natural Science Foundation[L182061] ; Bureau of International Cooperation of Chinese Academy of Sciences[173211KYSB20160053] ; Instrument Developing Project of the Chinese Academy of Sciences[YZ201502] ; Youth Innovation Promotion Association CAS[2017175] ; Natural Science Foundation of Heilongjiang Province[F201216]
WOS关键词	CENTRAL-NERVOUS-SYSTEM ; CLASSIFICATION ; SEGMENTATION ; TUMORS ; MRI
WOS研究方向	Radiology, Nuclear Medicine & Medical Imaging
语种	英语
出版者	ELSEVIER IRELAND LTD
WOS记录号	WOS:000469325700018
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China ; Beijing Natural Science Foundation ; Bureau of International Cooperation of Chinese Academy of Sciences ; Instrument Developing Project of the Chinese Academy of Sciences ; Youth Innovation Promotion Association CAS ; Natural Science Foundation of Heilongjiang Province
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/24380]
专题	中国科学院自动化研究所
通讯作者	Fang, Xiangming; Chen, Xuzhu; Tian, Jie
作者单位	1.Harbin Univ Sci & Technol, Sch Automat, Harbin 150080, Heilongjiang, Peoples R China 2.Chinese Acad Sci, Inst Automat, CAS Key Lab Mol Imaging, Beijing 100190, Peoples R China 3.Beihang Univ, Beijing Adv Innovat Ctr Big Data Based Precis Med, Sch Med, Beijing 100191, Peoples R China 4.Northeastern Univ, Sino Dutch Biomed & Informat Engn Sch, Shenyang 110169, Liaoning, Peoples R China 5.Univ Chinese Acad Sci, Beijing 100080, Peoples R China 6.Nanjing Med Univ, Wuxi Peoples Hosp, Imaging Ctr, 299 Qingyang Rd, Wuxi 214000, Jiangsu, Peoples R China 7.Capital Med Univ, Beijing Tiantan Hosp, Dept Radiol, 119 Nansihuan Xilu, Beijing 100050, Peoples R China 8.Xidian Univ, Sch Life Sci & Technol, Minist Educ, Engn Res Ctr Mol & Neuro Imaging, Xian 710126, Shaanxi, Peoples R China
推荐引用方式 GB/T 7714	Zhu, Yongbei,Man, Chuntao,Gong, Lixin,et al. A deep learning radiomics model for preoperative grading in meningioma[J]. EUROPEAN JOURNAL OF RADIOLOGY,2019,116:128-134.
APA	Zhu, Yongbei.,Man, Chuntao.,Gong, Lixin.,Dong, Di.,Yu, Xinyi.,...&Tian, Jie.(2019).A deep learning radiomics model for preoperative grading in meningioma.EUROPEAN JOURNAL OF RADIOLOGY,116,128-134.
MLA	Zhu, Yongbei,et al."A deep learning radiomics model for preoperative grading in meningioma".EUROPEAN JOURNAL OF RADIOLOGY 116(2019):128-134.