中国卒中杂志 ›› 2020, Vol. 15 ›› Issue (03): 283-289.DOI: 10.3969/j.issn.1673-5765.2020.03.010
李子孝,刘涛,丁玲玲,刘子阳,李鑫鑫,王拥军
收稿日期:
2020-01-10
出版日期:
2020-03-20
发布日期:
2020-03-20
通讯作者:
王拥军 yongjunwang111@aliyun.com
基金资助:
科技部“十三五”重点研发计划(2017YFC1310901,2016YFC0901002,2017YFC1307905,2015BAI12B00)
中国医学科学院脑血管病人工智能研究创新单元(2019RU018)
北京市科学技术委员会基于人工智能的脑血管病临床诊疗决策研究(Z201100005620010)
北京市百千万人才项目(2018A13)
北京市青年拔尖人才项目(2018000021223ZK03)
Received:
2020-01-10
Online:
2020-03-20
Published:
2020-03-20
李子孝,刘涛,丁玲玲,刘子阳,李鑫鑫,王拥军. 机器学习在脑血管病诊疗应用中的研究进展[J]. 中国卒中杂志, 2020, 15(03): 283-289.
LI Zi-Xiao, LIU Tao, DING Ling-Ling, LIU Zi-Yang, LI Xin-Xin, WANG Yong-Jun. Machine Learning in Stroke Care[J]. Chinese Journal of Stroke, 2020, 15(03): 283-289.
[1] WU S,WU B,LIU M,et al. Stroke in China:advances and challenges in epidemiology,prevention,and management[J]. Lancet Neurol,2019,18(4):394-405.[2] ZHOU M,WANG H,ZENG X,et al. Mortality,morbidity,and risk factors in China and its provinces,1990-2017:a systematic analysis for the GlobalBurden of Disease Study 2017[J]. Lancet,2019,394(10204):1145-1158.[3] POWERS W J,RABINSTEIN A A,ACKERSONT,et al. Guidelines for the early management ofpatients with acute ischemic stroke:2019 updateto the 2018 guidelines for the early management ofacute ischemic stroke:a guideline for healthcareprofessionals from the American Heart Association/American Stroke Association[J/OL]. Stroke,2019,50(12):e344-e418[2020-01-10]. https://doi.org/10.1161/STR.0000000000000211.[4] LI Z,SINGHAL A B,WANG Y. Stroke physiciantraining in China[J/OL]. Stroke,2017,48(12):e338-e340[2020-01-10]. https://doi.org/10.1161/STROKEAHA.117.019462.[5] 乔红艳,郭邦俊,张龙江. 机器学习在心血管影像中的研究进展[J]. 中华医学杂志,2019,99(17):1353-1357.[6] 卢光明,许强,张志强. 神经系统疾病人工智能医学影像现状与发展[J]. 中华放射学杂志,2018,52(10):734-737.[7] SONG T. Generative model-based ischemic strokelesion segmentation[DB/OL]. Electrical Engineeringand Systems Science,2019[2020-01-10]. https://arxiv.org/abs/1906.02392.[8] WU O,WINZECK S,GIESE A-K,et al. Big dataapproaches to phenotyping acute ischemic strokeusing automated lesion segmentation of multi-centermagnetic resonance imaging data[J]. Stroke,2019,50(7):1734-1741.[9] WOO I,LEE A,JUNG S C,et al. Fully automaticsegmentation of acute ischemic lesions ondiffusion-weighted imaging using convolutionalneural networks:comparison with conventionalalgorithms[J]. Korean J Radiol,2019,20(8):1275-1284.[10] KIM Y C,LEE J E,YU I,et al. Evaluation ofdiffusion lesion volume measurements in acuteischemic stroke using encoder-decoder convolutionalnetwork[J]. Stroke,2019,50(6):1444-1451. [11] WINZECK S,MOCKING S J T,BEZERRA R,et al. Ensemble of convolutional neural networksimproves automated segmentation of acute ischemiclesions using multiparametric diffusion-weightedMRI[J]. AJNR Am J Neuroradiol,2019,40(6):938-945.[12] BOLDSEN J K,ENGEDAL T S,PEDRAZA S,etal. Better diffusion segmentation in acute ischemicstroke through automatic tree learning anomalysegmentation[J]. Front Neuroinform,2018,12:21.[13] ZHANG R,ZHAO L,LOU W,et al. Automaticsegmentation of acute ischemic stroke from DWIusing 3-D fully convolutional DenseNets[J]. IEEETrans Med Imaging,2018,37(9):2149-2160.[14] CHEN L,BENTLEY P,RUECKERT D. Fullyautomatic acute ischemic lesion segmentationin DWI using convolutional neural networks[J].Neuroimage Clin,2017,15:633-643.[15] LIEW S L,ANGLIN J M,BANKS N W,et al. Alarge,open source dataset of stroke anatomical brainimages and manual lesion segmentations[J]. Sci Data,2018,5:180011.[16] QI K,YANG H,LI C,et al. X-net:Brain strokelesion segmentation based on depthwise separableconvolution and long-range dependencies.International Conference on Medical ImageComputing and Computer-Assisted Intervention(MICCAI)2019[C/OL]. Springer,2019:247-255[2020-01-10]. https://doi.org/ 10.1007/978-3-030-32248-9_28.[17] ZHOU Y J,HUANG W J,DONG P,et al. D-UNet:a dimension-fusion U shape network for chronicstroke lesion segmentation[DB/OL]. ElectricalEngineering and Systems Science,2019[2020-01-10].https://arxiv.org/abs/1908.05104.[18] YANG H,HUANG W,QI K,et al. CLCINet:Cross-Level Fusion and context inferencenetworks for lesion segmentation of chronicstroke. International conference on Medical ImageComputing and Computer-Assisted Intervention(MICCAI)2019[DB/OL]. Springer,2019:266-274[2020-01-10]. https://arxiv.org/abs/1907.07008.[19] XUE Y,XIE M,FARHAT F G,et al. A fully3D multi-path convolutional neural network withfeature fusion and feature weighting for automaticlesion identification in brain MRI images[DB/OL]. Electrical Engineering and Systems Science,2019[2020-01-10]. https://arxiv.org/abs/1907.07807.[20] DHAR R,FALCONE G J,CHEN Y,et al. Deeplearning for automated measurement of hemorrhageand perihematomal edema in supratentorialintracerebral hemorrhage[J]. Stroke,2020,51(2):648-651.[21] DEBETTE S,MARKUS H S. The clinicalimportance of white matter hyperintensities on brainmagnetic resonance imaging:systematic review andmeta-analysis[J/OL]. BMJ,2010,341:c3666[2020-01-10]. https://doi.org/10.1136/bmj.c3666.[22] LAWRENCE A J,ZEESTRATEN E A,BENJAMINP,et al. Longitudinal decline in structuralnetworks predicts dementia in cerebral smallvessel disease[J/OL]. Neurology,2018,90(21):e1898-e1910[2020-01-10]. https://doi.org/10.1212/WNL.0000000000005551.[23] JIANG J,LIU T,ZHU W,et al. UBO Detector - Acluster-based,fully automated pipeline for extractingwhite matter hyperintensities[J]. Neuroimage,2018,174:539-549.[24] LI H,JIANG G,ZHANG J,et al. Fullyconvolutional network ensembles for white matterhyperintensities segmentation in MR images[J].Neuroimage,2018,183:650-665.[25] RONNEBERGER O,FISCHER P,BROX T. U-net:Convolutional networks for biomedical imagesegmentation[C/OL]//International Conference onMedical Image Computing and Computer-assistedIntervention(MICCAI)2015. Springer,2015:234-241[2020-01-10]. https://doi.org/10.1007/978-3-319-24574-4_28.[26] WANG Y,CATINDIG J A,HILAL S,et al. Multistagesegmentation of white matter hyperintensity,cortical and lacunar infarcts[J]. Neuroimage,2012,60(4):2379-2388.[27] HALLER S,VERNOOIJ M W,KUIJER J P A,etal. Cerebral microbleeds:imaging and clinicalsignificance[J]. Radiology,2018,287(1):11-28.[28] ZHANG Y D,ZHANG Y,HOU X X,et al.Seven-layer deep neural network based on sparseautoencoder for voxelwise detection of cerebralmicrobleed[J]. Multimed Tools Appl,2018,77(9):10521-10538.[29] BRIDLE J S. Probabilistic interpretation offeedforward classification network outputs,withrelationships to statistical pattern recognition[J].Neurocomputing,1990,68:227-236.[30] QI DOU,HAO CHEN,LEQUAN YU,et al.Automatic detection of cerebral microbleeds fromMR Images via 3D convolutional neural networks[J].IEEE Trans Med Imaging,2016,35(5):1182-1195.[31] ZHU Y C,TZOURIO C,SOUMARE A,etal. Severity of dilated Virchow-Robin spaces isassociated with age,blood pressure,and MRI markers of small vessel disease:a population-basedstudy[J]. Stroke,2010,41(11):2483-2490.[32] GONZALEZ-CASTRO V,VALDES HERNANDEZM D C,CHAPPELL F M,et al. Reliability of anautomatic classifier for brain enlarged perivascularspaces burden and comparison with humanperformance[J]. Clin Sci(Lond),2017,131(13):1465-1481.[33] DUBOST F,YILMAZ P,ADAMS H,et al. Enlargedperivascular spaces in brain MRI:automatedquantification in four regions[J]. Neuroimage,2019,185:534-544.[34] CHILAMKURTHY S,GHOSH R,TANAMALAS,et al. Deep learning algorithms for detection ofcritical findings in head CT scans:a retrospectivestudy[J]. Lancet,2018,392(10162):2388-2396.[35] TITANO J J,BADGELEY M,SCHEFFLEIN J,etal. Automated deep-neural-network surveillance ofcranial images for acute neurologic events[J]. NatMed,2018,24(9):1337-1341.[36] LEE H,YUNE S,MANSOURI M,et al. Anexplainable deep-learning algorithm for the detectionof acute intracranial haemorrhage from smalldatasets[J]. Nat Biomed Eng,2019,3(3):173-182.[37] LIU J,XU H,CHEN Q,et al. Prediction ofhematoma expansion in spontaneous intracerebralhemorrhage using support vector machine[J].EBioMedicine,2019,43:454-459.[38] WANG H L,HSU W Y,LEE M H,et al. Automaticmachine-learning-based outcome prediction inpatients with primary intracerebral hemorrhage[J].Front Neurol,2019,10:910.[39] HEO J,YOON J G,PARK H,et al. Machinelearning-based model for prediction of outcomes inacute stroke[J]. Stroke,2019,50(5):1263-1265.[40] REHME A K,VOLZ L J,FEIS D L,et al.Identifying neuroimaging markers of motor disabilityin acute stroke by machine learning techniques[J].Cereb Cortex,2015,25(9):3046-3056.[41] SIEGEL J S,RAMSEY L E,SNYDER A Z,etal. Disruptions of network connectivity predictimpairment in multiple behavioral domains afterstroke[J/OL]. Proc Natl Acad Sci USA,2016,113(30):e4367-e4376[2020-01-10]. https://doi.org/10.1073/pnas.1521083113.[42] THOMALLA G,CHENG B,EBINGER M,etal. DWI-FLAIR mismatch for the identification ofpatients with acute ischaemic stroke within 4. 5 hof symptom onset(PRE-FLAIR):a multicentreobservational study[J]. Lancet Neurol,2011,10(11):978-986.[43] LEE H,LEE E-J,HAM S,et al. Machine Learningapproach to identify stroke within 4.5 hours[J/OL].Stroke,2020[2020-01-10]. https://doi.org/10.1161/STROKEAHA.119.027611.[44] CHUNG J W,KIM Y C,CHA J,et al.Characterization of clot composition in acute cerebralinfarct using machine learning techniques[J]. AnnClin Transl Neurol,2019,6(4):739-747.[45] BENTLEY P,GANESALINGAM J,CARLTONJONES A L,et al. Prediction of stroke thrombolysisoutcome using CT brain machine learning[J].Neuroimage Clin,2014,4:635-640.[46] QIU W,KUANG H,NAIR J,et al. Radiomicsbasedintracranial thrombus features on CT and CTApredict recanalization with intravenous alteplase inpatients with acute ischemic stroke[J]. AJNR Am JNeuroradiol,2019,40(1):39-44.[47] BACCHI S,ZERNER T,OAKDEN-RAYNER L,et al. Deep learning in the prediction of ischaemicstroke thrombolysis functional outcomes:apilot study[J/OL]. Acad Radiol,2020,27(2):e19-e23[2020-01-10]. https://doi.org/10.1016/j.acra.2019.03.015.[48] RODRIGUES G M,BARREIRA C,FROEHLERM,et al. Multicenter ALADIN:Automated LargeArtery Occlusion Detection in Stroke ImagingUsing Artificial Intelligence[J/OL]. Stroke,2019,50(Suppl_1):AWP71[2020-01-10]. https://www.ahajournals.org/doi/10.1161/str.50.suppl_1.WP71.[49] AMUKOTUWA S A,STRAKA M,SMITH H,etal. Automated detection of intracranial large vesselocclusions on computed tomography angiography:asingle center experience[J]. Stroke,2019,50(10):2790-2798.[50] ALBERS G W,WALD M J,MLYNASH M,et al.Automated calculation of Alberta Stroke Programearly CT score:validation in patients with largehemispheric infarct[J]. Stroke,2019,50(11):3277-3279. |
[1] | 蒲善宇, 周齐, 潘岳松. 我国脑血管病真实世界研究进展[J]. 中国卒中杂志, 2022, 17(12): 1289-1292. |
[2] | 冯玉婷, 刘芷含, 柴倩云, 罗慜婧, 高一城, 陶立元, 费宇彤. 真实世界研究中混杂因素和协变量控制——以脑血管病为例[J]. 中国卒中杂志, 2022, 17(12): 1304-1309. |
[3] | 金奥铭, 潘岳松. 倾向性评分法在脑血管病临床研究中的应用[J]. 中国卒中杂志, 2022, 17(12): 1310-1313. |
[4] | 蒋林, 李靖. 基于知识图谱分析人工智能用于卒中研究的现状及趋势 [J]. 中国卒中杂志, 2022, 17(12): 1372-1380. |
[5] | 仇鑫, 李子孝. DNA甲基化与脑血管病[J]. 中国卒中杂志, 2022, 17(11): 1159-1161. |
[6] | 邓宇含, 刘爽, 王子尧, 汪雨欣, 刘宝花. 基于结构化数据和机器学习模型预测普通人群卒中发病风险的系统评价和meta分析 [J]. 中国卒中杂志, 2022, 17(11): 1189-1197. |
[7] | 杨红娜, 冀瑞俊, 于凯, 何艳, 徐丽华, 王拥军. 缺血性脑血管病二级预防药物依从性及其影响因素研究 [J]. 中国卒中杂志, 2022, 17(11): 1227-1232. |
[8] | 索阅, 张喆, 谢雪微, 孔庆乐, 王拥军, 荆京. 7.0T MRI在脑血管病中的应用进展[J]. 中国卒中杂志, 2022, 17(11): 1257-1263. |
[9] | 胥智超, 李润芝, 米东华. 中枢胰岛素抵抗与脑血管病的关系[J]. 中国卒中杂志, 2022, 17(11): 1276-1282. |
[10] | 肖瑞珠, 李磊, 王孟, 姜勇, 郭敬鹏, 朱皞罡. 基于区块链的数字身份在脑血管病临床医疗数据应用的模式探索[J]. 中国卒中杂志, 2022, 17(10): 1089-1096. |
[11] | 罗秋雁, 黄文利, 冯慧宇, 陈玲, 周鸿雁. 脑室出血合并大面积脑梗死1例报道并文献复习[J]. 中国卒中杂志, 2022, 17(10): 1103-1107. |
[12] | 董漪, 张艳蓉, 曹文杰, 程忻, 董强, 顾宇翔. 浅谈脑血管病融合病房的基本概念和建设框架[J]. 中国卒中杂志, 2022, 17(10): 1139-1143. |
[13] | 陈虞, 王地, 李冠璋, 赵征, 李艺影. 脑血管病全血样本的采集、储存及转运标准化流程与解决方案[J]. 中国卒中杂志, 2022, 17(08): 911-914. |
[14] | 杨佳蕾, 陈思玎, 孟霞, 姜勇, 王拥军. 基于机器学习预测血糖异常急性缺血性卒中患者预后模型研究[J]. 中国卒中杂志, 2022, 17(07): 730-736. |
[15] | 宋新杰, 姜睿璇, 张怡君, 王安心, 李子孝, 赵性泉. 脑出血急性期不同血压参数与预后的关系[J]. 中国卒中杂志, 2022, 17(07): 769-773. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||