Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody

doi:10.3389/fneur.2022.947974

科研成果详情

题名	Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody
作者	Wang, Liang 1,2; Du, Lei 3; Li, Qinying 4; Li, Fang 2,5; Wang, Bei 6; Zhao, Yuanqi 7; Meng, Qiang 8; Li, Wenyu 9; Pan, Juyuan10; Xia, Junhui10; Wu, Shitao 11; Yang, Jie 12; Li, Heng 13; Ma, Jianhua 3; ZhangBao, Jingzi 1,2; Huang, Wenjuan 1,2; Chang, Xuechun 1,2; Tan, Hongmei 1,2; Yu, Jian 14; Zhou, Lei 1,2; Lu, Chuanzhen 1,2; Wang, Min 14; Dong, Qiang 1,2; Lu, Jiahong 1,2; Zhao, Chongbo 1,2; Quan, Chao 1,2
发表日期	2022-08-05
发表期刊	FRONTIERS IN NEUROLOGY 影响因子和分区
语种	英语
原始文献类型	Article
关键词	neuromyelitis optica spectrum disorder anti-aquaporin-4 antibody machine learning deep learning relapse prediction
其他关键词	SATRALIZUMAB ; EFFICACY
摘要	ObjectiveWe previously identified the independent predictors of recurrent relapse in neuromyelitis optica spectrum disorder (NMOSD) with anti-aquaporin-4 antibody (AQP4-ab) and designed a nomogram to estimate the 1- and 2-year relapse-free probability, using the Cox proportional hazard (Cox-PH) model, assuming that the risk of relapse had a linear correlation with clinical variables. However, whether the linear assumption fits real disease tragedy is unknown. We aimed to employ deep learning and machine learning to develop a novel prediction model of relapse in patients with NMOSD and compare the performance with the conventional Cox-PH model. MethodsThis retrospective cohort study included patients with NMOSD with AQP4-ab in 10 study centers. In this study, 1,135 treatment episodes from 358 patients in Huashan Hospital were employed as the training set while 213 treatment episodes from 92 patients in nine other research centers as the validation set. We compared five models with added variables of gender, AQP4-ab titer, previous attack under the same therapy, EDSS score at treatment initiation, maintenance therapy, age at treatment initiation, disease duration, the phenotype of the most recent attack, and annualized relapse rate (ARR) of the most recent year by concordance index (C-index): conventional Cox-PH, random survival forest (RSF), LogisticHazard, DeepHit, and DeepSurv. ResultsWhen including all variables, RSF outperformed the C-index in the training set (0.739), followed by DeepHit (0.737), LogisticHazard (0.722), DeepSurv (0.698), and Cox-PH (0.679) models. As for the validation set, the C-index of LogisticHazard outperformed the other models (0.718), followed by DeepHit (0.704), DeepSurv (0.698), RSF (0.685), and Cox-PH (0.651) models. Maintenance therapy was calculated to be the most important variable for relapse prediction. ConclusionThis study confirmed the superiority of deep learning to design a prediction model of relapse in patients with AQP4-ab-positive NMOSD, with the LogisticHazard model showing the best predictive power in validation.
资助项目	National Natural Science Foundation of China [82171341]; Shanghai Municipal Science and Technology Major Project [2018SHZDZX01]; ZHANGJIANG LAB; National Key Research and Development Program of China [2016YFC0901504]
出版者	FRONTIERS MEDIA SA
出版地	LAUSANNE
ISSN	1664-2295
卷号	13 页码:947974
DOI	10.3389/fneur.2022.947974
页数	10
WOS类目	Clinical Neurology ; Neurosciences
WOS研究方向	Neurosciences & Neurology
WOS记录号	WOS:000843944000001
收录类别	SCIE ; PUBMED ; SCOPUS
URL	查看原文
PubMed ID	35989911
SCOPUSEID	2-s2.0-85136493604
通讯作者地址	[Quan, Chao]Department of Neurology,Huashan Rare Disease Center,Huashan Hospital,Shanghai Medical College,Fudan University,Shanghai,China
Scopus学科分类	Neurology;Neurology (clinical)
引用统计	被引频次[WOS]：0 [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	https://kms.wmu.edu.cn/handle/3ETUA0LF/154270
专题	第一临床医学院（信息与工程学院）、附属第一医院_内科学_神经内科
通讯作者	Quan, Chao
作者单位	1.Department of Neurology,Huashan Rare Disease Center,Huashan Hospital,Shanghai Medical College,Fudan University,Shanghai,China; 2.National Center for Neurological Disorders (NCND),Shanghai,China; 3.Department of Neurology,The First Affiliated Hospital of Xinjiang Medical University,Xinjiang Uygur Autonomous Region,Urumqi,China; 4.Department of Rehabilitation Medicine,Jing'an District Centre Hospital of Shanghai,Fudan University,Shanghai,China; 5.Department of Rehabilitation Medicine,Huashan Hospital,Shanghai Medical College,Fudan University,Shanghai,China; 6.Department of Neurology,Jing'an District Centre Hospital of Shanghai,Fudan University,Shanghai,China; 7.Department of Neurology,The Second Affiliated Hospital of Guangzhou University of Chinese Medicine,Guangzhou,China; 8.Department of Neurology,The First People's Hospital of Yunnan Province,Kunming,China; 9.Department of Neurology,Sir Run Run Shaw Hospital,School of Medicine,Zhejiang University,Hangzhou,China; 10.Department of Neurology,The First Affiliated Hospital of Wenzhou Medical University,Wenzhou,China; 11.Department of Neurology,The Fifth Affiliated Hospital of Zhengzhou University,Zhengzhou,China; 12.Department of Neurology,Wuhan No.1 Hospital,Wuhan,China; 13.Department of Neurology,Central Hospital,Shandong First Medical University,Jinan,China; 14.Department of Ophthalmology and Vision Science,Eye and ENT Hospital,Shanghai Medical College,Fudan University,Shanghai,China
推荐引用方式 GB/T 7714	Wang, Liang,Du, Lei,Li, Qinying,et al. Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody[J]. FRONTIERS IN NEUROLOGY,2022,13:947974.
APA	Wang, Liang., Du, Lei., Li, Qinying., Li, Fang., Wang, Bei., ... & Quan, Chao. (2022). Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody. FRONTIERS IN NEUROLOGY, 13, 947974.
MLA	Wang, Liang,et al."Deep learning-based relapse prediction of neuromyelitis optica spectrum disorder with anti-aquaporin-4 antibody".FRONTIERS IN NEUROLOGY 13(2022):947974.