Share：
Share this content in WeChat
X
[Chinese] [PDF] 39 1
• Review •
Research progress of artificial intelligence in imaging diagnosis of ischemic stroke
JIANG Quan  LONG Xiaowu  WU Yuankui 

DOI:10.12015/issn.1674-8034.2026.02.025.


[Abstract] Ischemic stroke is characterized by high rates of disability and mortality, and early precise imaging evaluation is crucial for endovascular treatment decision-making and prognosis prediction. However, conventional imaging assessment methods have significant limitations. Non-contrast CT has low sensitivity in detecting hyperacute infarcts, while manual interpretation methods such as etiology classification and ASPECTS scoring are subjective, time-consuming, and poorly reproducible. Artificial intelligence (AI) technology offers a promising approach to address these challenges. Extensive research has been conducted on the value of imaging-based AI in key areas of ischemic stroke, with important progress achieved. This article reviews studies on the application of imaging AI in assisting ischemic stroke etiology classification, automated lesion identification and segmentation, quantitative analysis of infarct core and ischemic penumbra, and automation of ASPECTS scoring. It also discusses the limitations of current research and future directions, aiming to provide references for the development of AI-assisted diagnostic tools for ischemic stroke and to facilitate the establishment of a rapid, objective, and reproducible stroke imaging assessment process to improve patient outcomes.
[Keywords] artificial intelligence；stroke；ischemic stroke；magnetic resonance imaging；tomography, X-ray computed

JIANG Quan1   LONG Xiaowu2   WU Yuankui1*  

1 Department of Medical Imaging, Nanfang Hospital, Southern Medical University, Guangzhou 510515, China

2 Department of Radiology, Yunfu People's Hospital, Yunfu 527300, China

Corresponding author: WU Y K, E-mail: ripleyor@126.com

Conflicts of interest   None.

Received  2025-09-24
Accepted  2026-01-26
DOI: 10.12015/issn.1674-8034.2026.02.025
DOI:10.12015/issn.1674-8034.2026.02.025.

[1]
FEIGIN V L, ABATE M D, ABATE Y H, et al. Global, regional, and national burden of stroke and its risk factors, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021[J]. Lancet Neurol, 2024, 23(10): 973-1003. DOI: 10.1016/S1474-4422(24)00369-7.
[2]
HILKENS N A, CASOLLA B, LEUNG T W, et al. Stroke[J]. Lancet, 2024, 403(10446): 2820-2836. DOI: 10.1016/S0140-6736(24)00642-1.
[3]
RAJPURKAR P, LUNGREN M P. The current and future state of AI interpretation of medical images[J]. N Engl J Med, 2023, 388(21): 1981-1990. DOI: 10.1056/NEJMra2301725.
[4]
FERNANDES J N D, CARDOSO V E M, COMESAÑA-CAMPOS A, et al. Comprehensive review: machine and deep learning in brain stroke diagnosis[J/OL]. Sensors, 2024, 24(13): 4355 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/39001134/. DOI: 10.3390/s24134355.
[5]
JIANG B, PHAM N, VAN STAALDUINEN E K, et al. Deep learning applications in imaging of acute ischemic stroke: a systematic review and narrative summary[J/OL]. Radiology, 2025, 315(1): e240775 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/40197098/. DOI: 10.1148/radiol.240775.
[6]
YEDAVALLI V S, TONG E, MARTIN D, et al. Artificial intelligence in stroke imaging: Current and future perspectives[J/OL]. Clin Imag, 2021, 69: 246-254 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/32980785/. DOI: 10.1016/j.clinimag.2020.09.005.
[7]
BINSON V A, THOMAS S, SUBRAMONIAM M, et al. A review of machine learning algorithms for biomedical applications[J]. Ann Biomed Eng, 2024, 52(5): 1159-1183. DOI: 10.1007/s10439-024-03459-3.
[8]
YANG T H, SU Y Y, TSAI C L, et al. Magnetic resonance imaging-based deep learning imaging biomarker for predicting functional outcomes after acute ischemic stroke[J/OL]. Eur J Radiol, 2024, 174: 111405 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38447430/. DOI: 10.1016/j.ejrad.2024.111405.
[9]
CHRISTENSEN S, DEMEESTERE J, VERHAAREN B F J, et al. Semiautomated detection of early infarct signs on noncontrast CT improves interrater agreement[J]. Stroke, 2023, 54(12): 3090-3096. DOI: 10.1161/STROKEAHA.123.044058.
[10]
QIU W, KUANG H L, TELEG E, et al. Machine learning for detecting early infarction in acute stroke with non-contrast-enhanced CT[J]. Radiology, 2020, 294(3): 638-644. DOI: 10.1148/radiol.2020191193.
[11]
LU J, ZHOU Y R, LV W Z, et al. Identification of early invisible acute ischemic stroke in non-contrast computed tomography using two-stage deep-learning model[J]. Theranostics, 2022, 12(12): 5564-5573. DOI: 10.7150/thno.74125.
[12]
VAN HORN N, KNIEP H, BROOCKS G, et al. ASPECTS interobserver agreement of 100 investigators from the TENSION study[J]. Clin Neuroradiol, 2021, 31(4): 1093-1100. DOI: 10.1007/s00062-020-00988-x.
[13]
WEI J Y, SHANG K, WEI X E, et al. Deep learning-based automatic ASPECTS calculation can improve diagnosis efficiency in patients with acute ischemic stroke: a multicenter study[J]. Eur Radiol, 2025, 35(2): 627-639. DOI: 10.1007/s00330-024-10960-9.
[14]
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. DOI: 10.1007/s00234-018-2066-5.
[15]
JUAN C J, LIN S C, LI Y H, et al. Improving interobserver agreement and performance of deep learning models for segmenting acute ischemic stroke by combining DWI with optimized ADC thresholds[J]. Eur Radiol, 2022, 32(8): 5371-5381. DOI: 10.1007/s00330-022-08633-6.
[16]
GHEIBI Y, SHIRINI K, RAZAVI S N, et al. CNN-Res: deep learning framework for segmentation of acute ischemic stroke lesions on multimodal MRI images[J/OL]. BMC Med Inform Decis Mak, 2023, 23(1): 192 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/37752508/. DOI: 10.1186/s12911-023-02289-y.
[17]
OU Y L, HUANG S X, WONG K K, et al. BBox-Guided Segmentor: Leveraging expert knowledge for accurate stroke lesion segmentation using weakly supervised bounding box prior[J/OL]. Comput Med Imaging Graph, 2023, 107: 102236 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/37146318/. DOI: 10.1016/j.compmedimag.2023.102236.
[18]
LIU J M, LI J Y, WU Y F, et al. Deep learning-based segmentation of acute ischemic stroke MRI lesions and recurrence prediction within 1 year after discharge: a multicenter study[J/OL]. Neuroscience, 2025, 565: 222-231 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/39631660/. DOI: 10.1016/j.neuroscience.2024.12.002.
[19]
YUM K S, CHUNG J W, HA S Y, et al. A multicenter validation and calibration of automated software package for detecting anterior circulation large vessel occlusion on CT angiography[J/OL]. BMC Neurol, 2025, 25(1): 100 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/40065263/. DOI: 10.1186/s12883-025-04107-6.
[20]
BRUGNARA G, BAUMGARTNER M, SCHOLZE E D, et al. Deep-learning based detection of vessel occlusions on CT-angiography in patients with suspected acute ischemic stroke[J/OL]. Nat Commun, 2023, 14(1): 4938 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/37582829/. DOI: 10.1038/s41467-023-40564-8.
[21]
LUIJTEN S P R, WOLFF L, DUVEKOT M H C, et al. Diagnostic performance of an algorithm for automated large vessel occlusion detection on CT angiography[J]. J Neurointerv Surg, 2022, 14(8): 794-798. DOI: 10.1136/neurintsurg-2021-017842.
[22]
KUANG H L, LIU X Y, LIU J, et al. Large vessel occlusion identification network with vessel guidance and asymmetry learning on CT angiography of acute ischemic stroke patients[J/OL]. Med Image Anal, 2025, 101: 103490 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/39933335/. DOI: 10.1016/j.media.2025.103490.
[23]
KANG J, PARK J, KIM Y, et al. Deep-learning-based detection of large vessel occlusion: a comparison of CT and diffusion-weighted imaging[J/OL]. Digit Health, 2025, 11: 20552076251334040 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/40297357/. DOI: 10.1177/20552076251334040.
[24]
HAO M J, WANG L Y, XIONG Y Y. Neuroimaging evaluation of reperfusion therapy[J]. Chin J Stroke, 2024, 19(4): 459-467. DOI: 10.3969/j.issn.1673-5765.2024.04.009.
[25]
TOMASETTI L, HOLLESLI L J, ENGAN K, et al. Machine learning algorithms versus thresholding to segment ischemic regions in patients with acute ischemic stroke[J]. IEEE J Biomed Health Inform, 2022, 26(2): 660-672. DOI: 10.1109/JBHI.2021.3097591.
[26]
WANG C Y, SHI Z, YANG M, et al. Deep learning-based identification of acute ischemic core and deficit from non-contrast CT and CTA[J]. J Cereb Blood Flow Metab, 2021, 41(11): 3028-3038. DOI: 10.1177/0271678X211023660.
[27]
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/OL]. J Biomed Sci, 2020, 27(1): 80 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/32664906/. DOI: 10.1186/s12929-020-00672-9.
[28]
PENSATO U, DEMCHUK A M, MENON B K, et al. Cerebral infarct growth: pathophysiology, pragmatic assessment, and clinical implications[J]. Stroke, 2025, 56(1): 219-229. DOI: 10.1161/STROKEAHA.124.049013.
[29]
PINTO A, PEREIRA S, MEIER R, et al. Combining unsupervised and supervised learning for predicting the final stroke lesion[J/OL]. Med Image Anal, 2021, 69: 101888 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/33387909/. DOI: 10.1016/j.media.2020.101888.
[30]
NIELSEN A, HANSEN M B, TIETZE A, et al. Prediction of tissue outcome and assessment of treatment effect in acute ischemic stroke using deep learning[J]. Stroke, 2018, 49(6): 1394-1401. DOI: 10.1161/STROKEAHA.117.019740.
[31]
WOUTERS A, ROBBEN D, CHRISTENSEN S, et al. Prediction of stroke infarct growth rates by baseline perfusion imaging[J]. Stroke, 2022, 53(2): 569-577. DOI: 10.1161/STROKEAHA.121.034444.
[32]
ROBBEN D, BOERS A M M, MARQUERING H A, et al. Prediction of final infarct volume from native CT perfusion and treatment parameters using deep learning[J/OL]. Med Image Anal, 2020, 59: 101589 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/31683091/. DOI: 10.1016/j.media.2019.101589.
[33]
CZAP A L, SHETH S A. Overview of imaging modalities in stroke[J/OL]. Neurology, 2021, 97(20_Supplement_2) [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/34785603/. DOI: 10.1212/wnl.0000000000012794.
[34]
BERKHEMER O A, JANSEN I G H, BEUMER D, et al. Collateral status on baseline computed tomographic angiography and intra-arterial treatment effect in patients with proximal anterior circulation stroke[J]. Stroke, 2016, 47(3): 768-776. DOI: 10.1161/STROKEAHA.115.011788.
[35]
TAN D, LIU J Y, CHEN S X, et al. Automatic evaluating of multi-phase cranial CTA collateral circulation based on feature fusion attention network model[J]. IEEE Trans Nanobioscience, 2023, 22(4): 789-799. DOI: 10.1109/TNB.2023.3283049.
[36]
AKTAR M, XIAO Y M, TEHRANI A K Z, et al. SCANED: Siamese collateral assessment network for evaluation of collaterals from ischemic damage[J/OL]. Comput Med Imaging Graph, 2024, 113: 102346 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38364600/. DOI: 10.1016/j.compmedimag.2024.102346.
[37]
KIM Y C, CHUNG J W, BANG O Y, et al. A deep learning-based automatic collateral assessment in patients with acute ischemic stroke[J]. Transl Stroke Res, 2023, 14(1): 66-72. DOI: 10.1007/s12975-022-01036-1.
[38]
LE H L, ROH H G, KIM H J, et al. A 3D multi-task regression and ordinal regression deep neural network for collateral imaging from dynamic susceptibility contrast-enhanced MR perfusion in acute ischemic stroke[J/OL]. Comput Methods Programs Biomed, 2022, 225: 107071 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/35994873/. DOI: 10.1016/j.cmpb.2022.107071.
[39]
ADAMS H P, BENDIXEN B H, KAPPELLE L J, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment[J]. Stroke, 1993, 24(1): 35-41. DOI: 10.1161/01.str.24.1.35.
[40]
KIM B K, PARK S, HAN M K, et al. Deep learning for prediction of mechanism in acute ischemic stroke using brain diffusion magnetic resonance image[J]. J Neurocrit Care, 2023, 16(2): 85-93. DOI: 10.18700/jnc.230039.
[41]
RYU W S, SCHELLINGERHOUT D, LEE H, et al. Deep learning-based automatic classification of ischemic stroke subtype using diffusion-weighted images[J]. J Stroke, 2024, 26(2): 300-311. DOI: 10.5853/jos.2024.00535.
[42]
CHEN P H, GAO S, WANG Y J, et al. Classifying ischemic stroke, from TOAST to CISS[J]. CNS Neurosci Ther, 2012, 18(6): 452-456. DOI: 10.1111/j.1755-5949.2011.00292.x.
[43]
LI H X, LIU J, CHENG X Q, et al. Prediction of mixed ischemic stroke mechanism based on HR-MRI radiomics of intracranial arterial plaque[J]. Chin J Magn Reson Imaging, 2023, 14(3): 6-11, 27. DOI: 10.12015/issn.1674-8034.2023.03.002.
[44]
YIN X L, ZHAO Y S, HUANG F P, et al. Machine learning-based classification of anterior circulation cerebral infarction using computational fluid dynamics and CT perfusion metrics[J/OL]. Brain Sci, 2025, 15(4): 399 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/40309891/. DOI: 10.3390/brainsci15040399.
[45]
YU X, PAN J J, ZHAO X Y, et al. Risk and benefit evaluation: application of multiphase computed tomography angiography in mechanical thrombectomy for patients with acute ischemic stroke[J]. J Comput Assist Tomogr, 2021, 45(5): 736-742. DOI: 10.1097/RCT.0000000000001219.
[46]
KUANG Y Y, ZHANG L T, YE K L, et al. Clinical and imaging predictors for hemorrhagic transformation of acute ischemic stroke after endovascular thrombectomy[J]. J Neuroimaging, 2024, 34(3): 339-347. DOI: 10.1111/jon.13191.
[47]
MA Y, XU D Y, LIU Q, et al. Nomogram prediction model for the risk of intracranial hemorrhagic transformation after intravenous thrombolysis in patients with acute ischemic stroke[J/OL]. Front Neurol, 2024, 15: 1361035 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38515444/. DOI: 10.3389/fneur.2024.1361035.
[48]
MIAO L Q, PENG M Y, WANG T X, et al. The study of MRI radiomics and machine learning in the prediction of hemorrhagic transformation in acute stroke[J]. Chin J Magn Reson Imaging, 2022, 13(3): 18-21, 75. DOI: 10.12015/issn.1674-8034.2022.03.004.
[49]
JIANG L, ZHOU L L, YONG W, et al. A deep learning-based model for prediction of hemorrhagic transformation after stroke[J/OL]. Brain Pathol, 2023, 33(2): e13023 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/34608705/. DOI: 10.1111/bpa.13023.
[50]
ZHANG Y, XIE G, ZHANG L F, et al. Constructing machine learning models based on non-contrast CT radiomics to predict hemorrhagic transformation after stoke: a two-center study[J/OL]. Front Neurol, 2024, 15: 1413795 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/39286806/. DOI: 10.3389/fneur.2024.1413795.
[51]
RU X S, ZHAO S L, CHEN W D, et al. A weakly supervised deep learning model integrating noncontrasted computed tomography images and clinical factors facilitates haemorrhagic transformation prediction after intravenous thrombolysis in acute ischaemic stroke patients[J/OL]. Biomed Eng Online, 2023, 22(1): 129 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38115029/. DOI: 10.1186/s12938-023-01193-w.
[52]
REN H H, SONG H J, CUI S G, et al. Deep learning of noncontrast CT for fast prediction of hemorrhagic transformation of acute ischemic stroke: a multicenter study[J/OL]. Eur Radiol Exp, 2025, 9(1): 8 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/39812734/. DOI: 10.1186/s41747-024-00535-0.
[53]
HEO J, YOON Y, HAN H J, et al. Prediction of cerebral hemorrhagic transformation after thrombectomy using a deep learning of dual-energy CT[J]. Eur Radiol, 2024, 34(6): 3840-3848. DOI: 10.1007/s00330-023-10432-6.
[54]
KÜRTEN S, MUNOZ C, BESEOGLU K, et al. Decompressive hemicraniectomy for malignant middle cerebral artery infarction including patients with additional involvement of the anterior and/or posterior cerebral artery territory-outcome analysis and definition of prognostic factors[J]. Acta Neurochir, 2018, 160(1): 83-89. DOI: 10.1007/s00701-017-3329-3.
[55]
FOROUSHANI H M, HAMZEHLOO A, KUMAR A, et al. Accelerating prediction of malignant cerebral edema after ischemic stroke with automated image analysis and explainable neural networks[J]. Neurocrit Care, 2022, 36(2): 471-482. DOI: 10.1007/s12028-021-01325-x.
[56]
ZENG W X, LI W, HUANG K B, et al. Predicting futile recanalization, malignant cerebral edema, and cerebral herniation using intelligible ensemble machine learning following mechanical thrombectomy for acute ischemic stroke[J/OL]. Front Neurol, 2022, 13: 982783 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/36247767/. DOI: 10.3389/fneur.2022.982783.
[57]
WEN X H, HU X F, XIAO Y N, et al. Radiomics analysis for predicting malignant cerebral edema in patients undergoing endovascular treatment for acute ischemic stroke[J]. Diagn Interv Radiol, 2023, 29(2): 402-409. DOI: 10.4274/dir.2023.221764.
[58]
JIANG Q M, YU S, DONG X F, et al. Predictors and dynamic nomogram to determine the individual risk of malignant brain edema after endovascular thrombectomy in acute ischemic stroke[J]. J Clin Neurol, 2022, 18(3): 298-307. DOI: 10.3988/jcn.2022.18.3.298.
[59]
CUI J Z, YANG J Y, WANG Y, et al. Automatic segmentation of hemispheric CSF on MRI using deep learning: Quantifying cerebral edema following large hemispheric infarction[J/OL]. Heliyon, 2024, 10(5): e26673 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38463867/. DOI: 10.1016/j.heliyon.2024.e26673.
[60]
DROZDOWSKA B A, SINGH S, QUINN T J. Thinking about the future: a review of prognostic scales used in acute stroke[J/OL]. Front Neurol, 2019, 10: 274 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/30949127/. DOI: 10.3389/fneur.2019.00274.
[61]
DAIDONE M, FERRANTELLI S, TUTTOLOMONDO A. Machine learning applications in stroke medicine: advancements, challenges, and future prospectives[J]. Neural Regen Res, 2024, 19(4): 769-773. DOI: 10.4103/1673-5374.382228.
[62]
BORSOS B, ALLAART C G, VAN HALTEREN A. Predicting stroke outcome: a case for multimodal deep learning methods with tabular and CT Perfusion data[J/OL]. Artif Intell Med, 2024, 147: 102719 [2025-09-23]. https://pubmed.ncbi.nlm.nih.gov/38184355/. DOI: 10.1016/j.artmed.2023.102719.
[63]
MOULTON E, VALABREGUE R, PIOTIN M, et al. Interpretable deep learning for the prognosis of long-term functional outcome post-stroke using acute diffusion weighted imaging[J]. J Cereb Blood Flow Metab, 2023, 43(2): 198-209. DOI: 10.1177/0271678X221129230.

PREV Advances in magnetic resonance imaging for the evaluation of cerebral watershed infarction
NEXT Advances in predicting response to neoadjuvant therapy for hormone receptor-positive breast cancer based on multi-omics integration of DCE-MRI deep learning and tumor microenvironment
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn