The Role of Artificial Intelligence in Acute Stroke Imaging: Current Status and Future Directions

Abstract

Acute stroke imaging is paramount for quick diagnostic decisions using imaging modalities, including computerized tomography and magnetic resonance imaging. These modalities provide valuable information to determine the disease etiology and course of action. “Time is brain,” and as such, there are unique challenges regarding acute stroke imaging, including acute decision making, limited resource availability, and compromised image quality. Artificial intelligence (AI) tools may provide support for acute stroke imaging in clinical care settings and have the potential to mitigate many of these challenges. This paper investigates the role of AI in acute stroke imaging in research and industry, including reviewing 39 papers published between 2022 and 2025 that investigated the development and application of tools specific to interventional radiology and stroke and examined their tools' efficacy and the studies' consideration of data privacy, reproducibility, and practical usability. We also investigated four commercially available AI tools available for clinical use, focusing on their primary objectives, strengths, and limitations. We found that while AI tools demonstrate the potential for improving time-to-treatment and diagnostic accuracy, there are key limitations related to low reproducibility, the development of impractical tools, and minimal documentation about AI development and employment.

Keywords

AI - acute ischemic stroke - imaging - MRI - CT

Publikationsverlauf

Artikel online veröffentlicht:
06. November 2025

• References

• 1 Ospel JM, Holodinsky JK, Goyal M. Management of acute ischemic stroke due to large-vessel occlusion: JACC focus seminar. J Am Coll Cardiol 2020; 75 (15) 1832-1843
• 2 MICCAI. MICCAI Registered Challenges. Accessed June 13, 2025 at: https://miccai.org/index.php/special-interest-groups/challenges/miccai-registered-challenges/
• 3 Maier O, Menze BH, von der Gablentz J. et al. ISLES 2015 - a public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI. Med Image Anal 2017; 35: 250-269
• 4 Liu CF, Hsu J, Xu X. et al. Digital 3D brain MRI arterial territories atlas. Sci Data 2023; 10 (01) 74
• 5 Ding J, Zhou R, Fang X. et al. An image registration-based self-supervised Su-Net for carotid plaque ultrasound image segmentation. Comput Methods Programs Biomed 2024; 244: 107957
• 6 Gauriau R, Bizzo BC, Comeau DS. et al. Head CT deep learning model is highly accurate for early infarct estimation. Sci Rep 2023; 13 (01) 189
• 7 Lee KY, Liu CC, Chen DYT, Weng CL, Chiu HW, Chiang CH. Automatic detection and vascular territory classification of hyperacute staged ischemic stroke on diffusion weighted image using convolutional neural networks. Sci Rep 2023; 13 (01) 404
• 8 Aktar M, Tampieri D, Xiao Y, Rivaz H, Kersten-Oertel M. CASCADE-FSL: few-shot learning for collateral evaluation in ischemic stroke. Comput Med Imaging Graph 2025; 123: 102550
• 9 Ostmeier S, Axelrod B, Verhaaren BFJ. et al. Non-inferiority of deep learning ischemic stroke segmentation on non-contrast CT within 16-hours compared to expert neuroradiologists. Sci Rep 2023; 13 (01) 16153
• 10 Gómez S, Rangel E, Mantilla D. et al. APIS: a paired CT-MRI dataset for ischemic stroke segmentation - methods and challenges. Sci Rep 2024; 14 (01) 20543
• 11 Zavaliangos-Petropulu A, Tubi MA, Haddad E. et al. Testing a convolutional neural network-based hippocampal segmentation method in a stroke population. Hum Brain Mapp 2022; 43 (01) 234-243
• 12 Cao Z, Xu J, Song B. et al. Deep learning derived automated ASPECTS on non-contrast CT scans of acute ischemic stroke patients. Hum Brain Mapp 2022; 43 (10) 3023-3036
• 13 Liu G, Wu J, Dang C. et al. Machine learning for predicting motor improvement after acute subcortical infarction using baseline whole brain volumes. Neurorehabil Neural Repair 2022; 36 (01) 38-48
• 14 Williamson BJ, Khandwala V, Wang D. et al. Automated grading of enlarged perivascular spaces in clinical imaging data of an acute stroke cohort using an interpretable, 3D deep learning framework. Sci Rep 2022; 12 (01) 788
• 15 El-Hariri H, Souto Maior Neto LA, Cimflova P. et al. Evaluating nnU-Net for early ischemic change segmentation on non-contrast computed tomography in patients with acute ischemic stroke. Comput Biol Med 2022; 141: 105033
• 16 Sahoo PK, Mohapatra S, Wu CY, Huang KL, Chang TY, Lee TH. Automatic identification of early ischemic lesions on non-contrast CT with deep learning approach. Sci Rep 2022; 12 (01) 18054
• 17 Scavasine VC, Ferreti LA, da Costa RT. et al. Automated evaluation of ASPECTS from brain computerized tomography of patients with acute ischemic stroke. J Neuroimaging 2023; 33 (01) 134-137
• 18 Lee S, Lee E, Lee KS, Pyun SB. Explainable artificial intelligence on safe balance and its major determinants in stroke patients. Sci Rep 2024; 14 (01) 23735
• 19 Kim PJ, Kim D, Lee J. et al. Deep learning-based classification of diffusion-weighted imaging-fluid-attenuated inversion recovery mismatch. Sci Rep 2025; 15 (01) 5924
• 20 Kim PE, Yang H, Kim D. et al. Automated prediction of proximal middle cerebral artery occlusions in noncontrast brain computed tomography. Stroke 2024; 55 (06) 1609-1618
• 21 Liu Y, Yu Y, Ouyang J. et al. Functional outcome prediction in acute ischemic stroke using a fused imaging and clinical deep learning model. Stroke 2023; 54 (09) 2316-2327
• 22 Werdiger F, Visser M, Bivard A. et al. Benchmark dataset for clot detection in ischemic stroke vessel-based imaging: CODEC-IV. Neuroimage 2023; 271: 119985
• 23 Wouters A, Robben D, Christensen S. et al. Prediction of stroke infarct growth rates by baseline perfusion imaging. Stroke 2022; 53 (02) 569-577
• 24 Sun K, Shi R, Yu X. et al. Noninvasive imaging biomarker reveals invisible microscopic variation in acute ischaemic stroke (≤ 24 h): a multicentre retrospective study. Sci Rep 2025; 15 (01) 3743
• 25 Soleimani P, Farezi N. Utilizing deep learning via the 3D U-net neural network for the delineation of brain stroke lesions in MRI image. Sci Rep 2023; 13 (01) 19808
• 26 Rachmadi MF, Valdés-Hernández MDC, Makin S, Wardlaw J, Skibbe H. Prediction of white matter hyperintensities evolution one-year post-stroke from a single-point brain MRI and stroke lesions information. Sci Rep 2025; 15 (01) 1208
• 27 Hong SW, Song HN, Choi JU. et al. Automated in-depth cerebral arterial labelling using cerebrovascular vasculature reframing and deep neural networks. Sci Rep 2023; 13 (01) 3255
• 28 Ishikita A, McIntosh C, Hanneman K. et al. Machine learning for prediction of adverse cardiovascular events in adults with repaired tetralogy of Fallot using clinical and cardiovascular magnetic resonance imaging variables. Circ Cardiovasc Imaging 2023; 16 (06) e015205
• 29 Jianbo C, Hanqi P, Yihao C. et al. Weakly supervised multitask learning models to identify symptom onset time of unclear-onset intracerebral hemorrhage. Int J Stroke 2022; 17 (07) 785-792
• 30 Jiang L, Zhou L, Yong W. et al. A deep learning-based model for prediction of hemorrhagic transformation after stroke. Brain Pathol 2023; 33 (02) e13023
• 31 Mair G, White P, Bath PM. et al. External Validation of e-ASPECTS software for interpreting brain CT in stroke. Ann Neurol 2022; 92 (06) 943-957
• 32 Mohapatra S, Lee TH, Sahoo PK, Wu CY. Localization of early infarction on non-contrast CT images in acute ischemic stroke with deep learning approach. Sci Rep 2023; 13 (01) 19442
• 33 Ozkara BB, Karabacak M, Hoseinyazdi M. et al. Utilizing imaging parameters for functional outcome prediction in acute ischemic stroke: a machine learning study. J Neuroimaging 2024; 34 (03) 356-365
• 34 Aktar M, Reyes J, Tampieri D, Rivaz H, Xiao Y, Kersten-Oertel M. Deep learning for collateral evaluation in ischemic stroke with imbalanced data. Int J Comput Assist Radiol Surg 2023; 18 (04) 733-740
• 35 Olivot JM, Mosimann PJ, Labreuche J. et al. Impact of diffusion-weighted imaging lesion volume on the success of endovascular reperfusion therapy. Stroke 2013; 44 (08) 2205-2211
• 36 Albers GW, Thijs VN, Wechsler L. et al. Magnetic resonance imaging profiles predict clinical response to early reperfusion: the diffusion and perfusion imaging evaluation for understanding stroke evolution (DEFUSE) study. Ann Neurol 2006; 60 (05) 508-517
• 37 Yoo AJ, Verduzco LA, Schaefer PW, Hirsch JA, Rabinov JD, González RG. MRI-based selection for intra-arterial stroke therapy: value of pretreatment diffusion-weighted imaging lesion volume in selecting patients with acute stroke who will benefit from early recanalization. Stroke 2009; 40 (06) 2046-2054
• 38 Goyal M, Ospel JM, Menon B. et al. Challenging the ischemic core concept in acute ischemic stroke imaging. Stroke 2020; 51 (10) 3147-3155
• 39 Barber PA, Demchuk AM, Zhang J, Buchan AM. Validity and reliability of a quantitative computed tomography score in predicting outcome of hyperacute stroke before thrombolytic therapy. ASPECTS Study Group. Alberta Stroke Programme Early CT Score. Lancet 2000; 355 (9216): 1670-1674
• 40 Gupta AC, Schaefer PW, Chaudhry ZA. et al. Interobserver reliability of baseline noncontrast CT Alberta Stroke Program Early CT Score for intra-arterial stroke treatment selection. AJNR Am J Neuroradiol 2012; 33 (06) 1046-1049
• 41 Deng J, Dong W, Socher R, Li LJ, Li K, Li F-F. ImageNet: A large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE; 2009: 248-255
• 42 Jansen IGH, Mulder MJHL, Goldhoorn RB. MR CLEAN Registry investigators. Endovascular treatment for acute ischaemic stroke in routine clinical practice: prospective, observational cohort study (MR CLEAN Registry). BMJ 2018; 360: k949
• 43 He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2016: 770-778
• 44 Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In: 2015: 234-241
• 45 Soun JE, Zolyan A, McLouth J. et al. Impact of an automated large vessel occlusion detection tool on clinical workflow and patient outcomes. Front Neurol 2023; 14: 1179250
• 46 Simonyan K, Zisserman A. . Very deep convolutional networks for large-scale image recognition. 3rd International Conference on Learning Representations. April 10, 2015. San Diego, CA
• 47 Szegedy C, Liu W, Jia Y. et al. Going deeper with convolutions. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2015: 1-9
• 48 Czap AL, Bahr-Hosseini M, Singh N. et al. Machine learning automated detection of large vessel occlusion from mobile stroke unit computed tomography angiography. Stroke 2022; 53 (05) 1651-1656
• 49 Polson JS, Zhang H, Nael K. et al. Identifying acute ischemic stroke patients within the thrombolytic treatment window using deep learning. J Neuroimaging 2022; 32 (06) 1153-1160
• 50 Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely Connected Convolutional Networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2017: 2261-2269
• 51 Isensee F, Jaeger PF, Kohl SAA, Petersen J, Maier-Hein KH. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 2021; 18 (02) 203-211
• 52 Sun J, Li Q, Liu Y. et al. Pathological asymmetry-guided progressive learning for acute ischemic stroke infarct segmentation. IEEE Trans Med Imaging 2024; 43 (12) 4146-4160
• 53 Ahmed R, Al Shehhi A, Werghi N, Seghier ML. Segmentation of stroke lesions using transformers-augmented MRI analysis. Hum Brain Mapp 2024; 45 (11) e26803
• 54 Breiman L. Random forests. Mach Learn 2001; 45 (01) 5-32
• 55 Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A. CatBoost: unbiased boosting with categorical features. In: Bengio S, Wallach H, Larochelle H, Grauman K, Cesa-Bianchi N, Garnett R. eds. Advances in Neural Information Processing Systems. Vol 31. Curran Associates, Inc.; 2018. . Accessed October 24, 2025 at: https://proceedings.neurips.cc/paper_files/paper/2018/file/14491b756b3a51daac41c24863285549-Paper.pdf
• 56 Chen T, Guestrin C. XGBoost. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM; 2016: 785-794
• 57 Heo J, Yoo J, Lee H. et al. Prediction of hidden coronary artery disease using machine learning in patients with acute ischemic stroke. Neurology 2022; 99 (01) e55-e65
• 58 Gong B, Khalvati F, Ertl-Wagner BB, Patlas MN. Artificial intelligence in emergency neuroradiology: current applications and perspectives. Diagn Interv Imaging 2025; 106 (04) 135-142
• 59 Temmen SE, Becks MJ, Schalekamp S, van Leeuwen KG, Meijer FJA. Duration and accuracy of automated stroke CT workflow with AI-supported intracranial large vessel occlusion detection. Sci Rep 2023; 13 (01) 12551
• 60 Azeem H, Kim HW, Ballekere AN. et al. Abstract TMP57: user engagement is associated with clinical impacts of automated LVO detection software. Stroke 2024; 55 (Suppl. 01)
• 61 Khunte M, Chae A, Wang R. et al. Trends in clinical validation and usage of US Food and Drug Administration-cleared artificial intelligence algorithms for medical imaging. Clin Radiol 2023; 78 (02) 123-129
• 62 Rapid AI. RapidAI Neurovascular Solutions. Accessed May 27, 2025 at: https://www.rapidai.com/neurovascular
• 63 Viz.ai. Viz. Accessed May 15, 2025 at: https://www.viz.ai/large-vessel-occlusion
• 64 Brainomix. Transforming Stroke Care Through Simple Imaging. Accessed May 16, 2025 at: https://www.brainomix.com/stroke/e-aspects/
• 65 Aidoc Medical Limited. The Next Generation of Neuro AI and Care Coordination. Accessed June 3, 2025 at: https://www.aidoc.com/solutions/neuro/
• 66 Willemink MJ, Noël PB. The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence. Eur Radiol 2019; 29 (05) 2185-2195
• 67 Herrmann DM, Kalafut JF, Bizzo CB. et al. inventors; GE Precision Healthcare LLC, Partners HealthCare System, Inc., The General Hospital Corporation, The Brigham and Women's Hospital, Inc., assignee. CTA Large Vessel Occlusion Model. US patent 11,751,832. September 12, 2023.
• 68 Canada's Drug Agency L'Agence des médicaments du Canada. RapidAI for Stroke Detection: Main Report. Canadian Journal of Health Technologies 2024; 4 (11) 26-29
• 69 Rapid AI. Review. Canadian Agency for Drugs and Technologies in Health. 2024
• 70 Slater L, Ravintharan N, Goergen S. et al. RapidAI compared with human readers of acute stroke imaging for detection of intracranial vessel occlusion. Stroke Vasc Intervent Neurol 2024 4. 02
• 71 Mansi C, Golan D, Levi G, Duchin Y. inventors; Viz.ai Inc., assignee. Method and System for Computer-Aided Triage of Stroke. US patent 10,902,602. January 26, 2021.
• 72 Devlin T, Gao L, Collins O. et al. VALIDATE—utilization of the Viz.ai mobile stroke care coordination platform to limit delays in LVO stroke diagnosis and endovascular treatment. Front Stroke 2024 3.
• 73 Figurelle ME, Meyer DM, Perrinez ES. et al. Viz.ai implementation of stroke augmented intelligence and communications platform to improve indicators and outcomes for a comprehensive stroke center and network. AJNR Am J Neuroradiol 2023; 44 (01) 47-53
• 74 Puviani L, Bou-Zeid W, Joly O. inventors; Brainomix Limited, assignee. Methods for determining the size and density of a lesion from a medical imaging scan. US patent application publication 20,250,166,215. May 22, 2025.
• 75 Brinjikji W, Abbasi M, Arnold C. et al. e-ASPECTS software improves interobserver agreement and accuracy of interpretation of aspects score. Interv Neuroradiol 2021; 27 (06) 781-787
• 76 Mallon D, Fallon M, Blana E. et al. Real-world evaluation of Brainomix e-Stroke software. Stroke Vasc Neurol 2024; 9 (05) 497-504
• 77 Nagaratnam K, Neuhaus A, Briggs JH. et al. Artificial intelligence-based decision support software to improve the efficacy of acute stroke pathway in the NHS: an observational study. Front Neurol 2024; 14: 1329643
• 78 Walach E, Walach E. et al. Inventors; AIDOC Medical Ltd., assignee. System for determining the presence of features in a dataset. US patent 12,014,489. June 18, 2024.
• 79 Bankston M, Ayub M, Sheikh M. et al. Abstract 175: a retrospective analysis comparing AIDoc and RAPIDAI in the detection of large vessel occlusions. Stroke Vasc Intervent Neurol 2024 4. (S1):
• 80 Olsen AL, Ginat DT. Assessment of commercially available artificial intelligence software for differentiating hemorrhage from contrast on head CT following thrombolysis for ischemic stroke. Front Neurol 2025; 16: 1458142
• 81 Dagher R, Ozkara BB, Karabacak M. et al. Artificial intelligence/machine learning for neuroimaging to predict hemorrhagic transformation: systematic review/meta-analysis. J Neuroimaging 2024; 34 (05) 505-514
• 82 Chen S, Sedghi Gamechi Z, Dubost F, van Tulder G, de Bruijne M. An end-to-end approach to segmentation in medical images with CNN and posterior-CRF. Med Image Anal 2022; 76: 102311
• 83 Vrudhula A, Kwan AC, Ouyang D, Cheng S. Machine learning and bias in medical imaging: opportunities and challenges. Circ Cardiovasc Imaging 2024; 17 (02) e015495
• 84 Wang Z, Yang Y, Chen Y. et al. Mutual information guided diffusion for zero-shot cross-modality medical image translation. IEEE Trans Med Imaging 2024; 43 (08) 2825-2838
• 85 Jeon YJ, Roh HG, Jung S. et al. Clinical feasibility of deep learning-driven magnetic resonance angiography collateral map in acute anterior circulation ischemic stroke. Sci Rep 2025; 15 (01) 2304
• 86 Wardlaw JM, Mair G, von Kummer R. et al. Accuracy of automated computer-aided diagnosis for stroke imaging: a critical evaluation of current evidence. Stroke 2022; 53 (07) 2393-2403