人工智能影像技术在卒中诊疗中的应用

doi:10.3969/j.issn.1673-5765.2021.07.003

摘要/Abstract

摘要： 多模态影像是全面评估缺血性卒中发生发展的重要手段，其数据信息较多，且分析难度较大，传统的计算机视觉方法依赖于手工提取特征，在复杂任务上性能有限且通用性不佳。人工智能影像技术主要指利用人工智能方法处理计算机视觉任务，包括图像分类，病灶定位、检测、分割等，可深入挖掘多维影像学信息，并综合其他临床资料，已在卒中的早期筛查、病灶识别、病情诊断和预后预测等方面展开了广泛而深入的研究工作。人工智能影像技术在快速精准的影像学分析及标准化诊疗辅助方面具有一定应用价值，但在临床价值验证和产品转化方面仍存在不足，且其在临床实践中的应用发展亦面临诸多挑战。

文章导读： 人工智能影像技术对于卒中诊疗可发挥积极作用，但该技术仍面临诸多挑战，未来还需继续深入研发针
对卒中的临床辅助决策系统，使更多患者受益。

关键词: 人工智能; 卒中; 影像

Abstract: Multi-modality imaging is an important means to evaluate the occurrence and development of ischemic stroke. For complex and amounts of information inside the imaging, the imaging data analysis has certain difficulty. The traditional computer vision methods rely on manual extraction of information, which made the performance limited and the generality poor in dealing with complex tasks. Artificial intelligence (AI) imaging technology mainly refers to using AI to deal with computer vision tasks, including image classification, lesions localization, detection, imaging segmentation and etc. AI imaging analysis software can capture multidimensional imaging information, so AI imaging, combing with clinical information, can be applicated in many fields such as early screening, lesions recognition, diagnosis, and prognosis prediction of stroke. The preliminary application and deep research in the above fields are being carried out. AI imaging technology has showed certain value in rapid and precise imaging analysis and assisting in standardized diagnosis and treatment of stroke, while it has some deficiency in clinical validation and translation. At present, AI imaging technology still faces many challenges in clinical application.

Key words: Artificial intelligence; Stroke; Medical imaging

吴韵阳, 高键东, 吴及. 人工智能影像技术在卒中诊疗中的应用[J]. 中国卒中杂志, 2021, 16(07): 651-656.

WU Yun-Yang, GAO Jian-Dong, WU Ji. Application of Artificial Intelligence Imaging in Diagnosis and Treatment of Stroke[J]. Chinese Journal of Stroke, 2021, 16(07): 651-656.

参考文献

[1] WANG W Z，JIANG B，SUN H X，et al. Prevalence，
incidence，and mortality of stroke in China：results
from a nationwide population-based survey of 480
687 adults[J]. Circulation，2017，135（8）：759-771.
[2] 王拥军，李子孝，丁玲玲. 人工智能在卒中诊疗的研
究和应用：曙光初现，任重道远[J]. 中国卒中杂志，
2020，15（3）：223-227.
[3] 彭斌，吴波. 中国急性缺血性卒中诊治指南2018[J].
中华神经科杂志，2018，51（9）：666-682.
[4] TANG F H，NG D K，CHOW D K. An image
feature approach for computer-aided detection of
ischemic stroke[J]. Comput Biol Med，2011，41（7）：
529-536.
[5] ABEDI V，GOYAL N，TSIVGOULIS G，et al.
Novel screening tool for stroke using artificial neural
network[J]. Stroke，2017，48（6）：1678-1681.
[6] CHEN L，BENTLEY P，RUECKERT D. Fully
automatic acute ischemic lesion segmentation in
DWI using convolutional neural networks[J/OL].
Neuroimage Clin，2017，15：633-643[2021-05-01].
https：//doi.org/10.1016/j.nicl.2017.06.016.
[7] GUERRERO R，QIN C，OKTAY O，et al. White
matter hyperintensity and stroke lesion segmentation
and differentiation using convolutional neural
networks[J/OL]. Neuroimage Clin，2018，17：918-
934[2021-05-01]. https：//doi.org/10.1016/j.nicl.2017.
12.022.
[8] ÖMAN O，MÄKELÄ T，SALLI E，et al. 3D
convolutional neural networks applied to CT
angiography in the detection of acute ischemic
stroke[J]. Eur Radiol Exp，2019，3（1）：1-11.
[9] TAKAHASHI N，LEE Y，TSAI D，et al. An
automated detection method for the MCA dot sign
of acute stroke in unenhanced CT[J]. Radiol Phys
Technol，2014，7（1）：79-88.
[10] CHEN Z C，ZHANG R T，XU F Z，et al. Novel
prehospital prediction model of large vessel
occlusion using artificial neural network[J/OL]. Front
Aging Neurosci，2018，10：181[2021-05-01].
https：//doi.org/10.3389/fnagi.2018.00181.
[11] MURRAY N M，UNBERATH M，HAGER G D，et
al. Artificial intelligence to diagnose ischemic stroke
and identify large vessel occlusions：a systematic
review[J]. J Neurointerv Surg，2020，12（2）：156-
164.
[12] CHATTERJEE A，SOMAYAJI N R，KABAKIS I
M. Abstract WMP16：artificial intelligence detection
of cerebrovascular large vessel occlusion-nine month，
650 patient evaluation of the diagnostic accuracy
and performance of the Viz. ai LVO algorithm[J/OL].
Stroke，2019，50（Suppl_1）：AWMP16[2021-05-01].
https：//doi.org/10.1161/str.50.suppl_1.WMP16.
[13] BARREIRA C，BOUSLAMA M，LIM J，et
al. E-108 Aladin study：automated large artery
occlusion detection in stroke imaging study–a
multicenter analysis[J/OL]. J Neurointerv Surg，2018，
10（Suppl 2）：A101-A102[2021-05-01]. https：//doi.
org/10.1136/neurintsurg-2018-SNIS.184.
[14] GOYAL M，DEMCHUK A M，MENON B K，et
al. Randomized assessment of rapid endovascular
treatment of ischemic stroke[J]. N Engl J Med，2015，
372（11）：1019-1030.
[15] NAGEL S，SINHA D，DAY D，et al. e-ASPECTS
software is non-inferior to neuroradiologists in
applying the ASPECT score to computed tomography
scans of acute ischemic stroke patients[J]. Int J
Stroke，2017，12（6）：615-622.
[16] GUBERINA N，DIETRICH U，RADBRUCH A，et
al. Detection of early infarction signs with machine
learning-based diagnosis by means of the Alberta
Stroke Program Early CT Score（ASPECTS）in the
clinical routine[J]. Neuroradiology，2018，60（9）：
889-901.
[17] KUANG H，NAJM M，CHAKRABORTY D，et al.
Automated ASPECTS on noncontrast CT scans in
patients with acute ischemic stroke using machine
learning[J]. AJNR Am J Neuroradiol，2019，40（1）：
33-38.
[18] GRUNWALD I Q，KULIKOVSKI J，REITH

W，et al. Collateral automation for triage in stroke：

evaluating automated scoring of collaterals in
acute stroke on computed tomography scans[J].
Cerebrovasc Dis，2019，47（5/6）：217-222.
[19] BIVARD A，LEVI C，SPRATT N，et al. Perfusion
CT in acute stroke：a comprehensive analysis of
infarct and penumbra[J]. Radiology，2013，267（2）：
543-550.
[20] VAGAL A，WINTERMARK M，NAEL K，et al.
Automated CT perfusion imaging for acute ischemic
stroke：pearls and pitfalls for real-world use[J].
Neurology，2019，93（20）：888-898.
[21] KOOPMAN M S，BERKHEMER O A，GEUSKENS
R R E G，et al. Comparison of three commonly used
CT perfusion software packages in patients with
acute ischemic stroke[J]. J Neurointerv Surg，2019，
11（12）：1249-1256.
[22] KUO D P，KUO P C，CHEN Y C，et al. Machine
learning-based segmentation of ischemic penumbra
by using diffusion tensor metrics in a rat model[J]. J
Biomed Sci，2020，27（1）：1-11.
[23] LEE H，LEE E J，HAM S，et al. Machine learning
approach to identify stroke within 4.5 hours[J].
Stroke，2020，51（3）：860-866.
[24] GARG R，OH E，NAIDECH A，et al. Automating
ischemic stroke subtype classification using machine
learning and natural language processing[J]. J Stroke
Cerebrovasc Dis，2019，28（7）：2045-2051.
[25] BENTLEY P，GANESALINGAM J，CARLTON
JONES A L，et al. Prediction of stroke thrombolysis
outcome using CT brain machine learning[J/OL].
Neuroimage Clin，2014，4：635-640[2021-05-01].
https：//doi.org/10.1016/j.nicl.2014.02.003.
[26] KIM B J，KIM Y H，KIM N，et al. Lesion locationbased
prediction of visual field improvement after
cerebral infarction[J/OL]. PLoS One，2015，10（11）：
e0143882[2021-05-01]. https：//doi.org/10.1371/
journal.pone.0143882.
[27] RONDINA J M，FILIPPONE M，GIROLAMI M，
et al. Decoding post-stroke motor function from
structural brain imaging[J/OL]. Neuroimage Clin，
2016，12：372-380[2021-05-01]. https：//doi.org/10.
1016/j.nicl.2016.07.014.
[28] HEO J，YOON J G，PARK H，et al. Machine
learning-based model for prediction of outcomes in
acute stroke[J]. Stroke，2019，50（5）：1263-1265.
[29] MONTEIRO M，FONSECA A C，FREITAS A T，et
al. Using machine learning to improve the prediction
of functional outcome in ischemic stroke patients[J].
IEEE/ACM Trans Comput Biol Bioinform，2018，15
（6）：1953-1959.
[30] VAN HORN N，KNIEP H，LEISCHNER H，et al.
Predictors of poor clinical outcome despite complete
reperfusion in acute ischemic stroke patients[J]. J
Neurointerv Surg，2021，13（1）：14-18.
[31] ÇELIK G，BAYKAN Ö K，KARA Y，et al.
Predicting 10-day mortality in patients with strokes
using neural networks and multivariate statistical
methods[J]. J Stroke Cerebrovasc Dis，2014，23（6）：
1506-1512.
[32] OERMANN E K，RUBINSTEYN A，DING
D，et al. Using a machine learning approach to
predict outcomes after radiosurgery for cerebral
arteriovenous malformations[J]. Sci Rep，2016，6
（1）：1-12.