BACKGROUND AND PURPOSE: Identifying the presence and extent of intracranial thrombi is crucial in selecting patients with acute ischemic stroke for treatment. This article aims to develop an automated approach to quantify thrombus on NCCT and CTA in patients with stroke. MATERIALS AND METHODS: A total of 499 patients with large-vessel occlusion from the Safety and Efficacy of Nerinetide in Subjects Undergoing Endovascular Thrombectomy for Stroke (ESCAPE-NA1) trial were included. All patients had thin-section NCCT and CTA images. Thrombi contoured manually were used as reference standard. A deep learning approach was developed to segment thrombi automatically. Of 499 patients, 263 and 66 patients were randomly selected to train and validate the deep learning model, respectively; the remaining 170 patients were independently used for testing. The deep learning model was quantitatively compared with the reference standard using the Dice coefficient and volumetric error. The proposed deep learning model was externally tested on 83 patients with and without large-vessel occlusion from another independent trial. RESULTS: The developed deep learning approach obtained a Dice coefficient of 70.7% (interquartile range, 58.0%-77.8%) in the internal cohort. The predicted thrombi length and volume were correlated with those of expert-contoured thrombi (r = 0.88 and 0.87, respectively; P < .001). When the derived deep learning model was applied to the external data set, the model obtained similar results in patients with large-vessel occlusion regarding the Dice coefficient (66.8%; interquartile range, 58.5%-74.6%), thrombus length (r = 0.73), and volume (r = 0.80). The model also obtained a sensitivity of 94.12% (32/34) and a specificity of 97.96% (48/49) in classifying large-vessel occlusion versus non-large-vessel occlusion. CONCLUSIONS: The proposed deep learning method can reliably detect and measure thrombi on NCCT and CTA in patients with acute ischemic stroke.

College of Electronic Engineering Xi'an Shiyou University Xi'an Shaanxi China

Department of Community Health Sciences

Department of Diagnostic and Interventional Neuroradiology University Hospital Hamburg Hamburg Germany

Department of Medicine and Department of Radiology Cumming School of Medicine University of Calgary Calgary Alberta Canada

Department of Neurology and Cerebrovascular Center Seoul National University Bundang Hospital Seongnam si Gyeonggi do Korea

From the Department of Clinical Neurosciences and Hotchkiss Brain Institute

School of Life Science and Technology Huazhong University of Science and Technology Wuhan Hubei China

St Anne's University Hospital Brno and Faculty of Medicine Masaryk University Brno Czech Republic

AJNR Am J Neuroradiol. 2023 Sep;44(9):E41. doi: 10.3174/ajnr.A7972. PubMed

Zobrazit více v PubMed

Goyal M, Menon BK, van Zwam WH, et al. ; HERMES Collaborators. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016;387:1723–31 10.1016/S0140-6736(16)00163-X PubMed DOI

Menon BK, Hill MD, Davalos A, et al. . Efficacy of endovascular thrombectomy in patients with M2 segment middle cerebral artery occlusions: meta-analysis of data from the HERMES Collaboration. J Neurointerv Surg 2019;11:1065–69 10.1136/neurintsurg-2018-014678 PubMed DOI

Menon BK, Al-Ajlan FS, Najm M, et al. ; INTERRSeCT Study Investigators. Association of clinical, imaging, and thrombus characteristics with recanalization of visible intracranial occlusion in patients with acute ischemic stroke. JAMA 2018;320:1017–26 10.1001/jama.2018.12498 PubMed DOI PMC

Riedel CH, Zimmermann P, Jensen-Kondering U, et al. . The importance of size: successful recanalization by intravenous thrombolysis in acute anterior stroke depends on thrombus length. Stroke 2011;42:1775–77 10.1161/STROKEAHA.110.609693 PubMed DOI

Qiu W, Kuang H, Nair J, et al. . Radiomics-based intracranial thrombus features on CT and CTA predict recanalization with intravenous alteplase in patients with acute ischemic stroke. AJNR Am J Neuroradiol 2019;40:39–44 10.3174/ajnr.A5918 PubMed DOI PMC

Hofmeister J, Bernava G, Rosi A, et al. . Clot-based radiomics predict a mechanical thrombectomy strategy for successful recanalization in acute ischemic stroke. Stroke 2020;51:2488–94 10.1161/STROKEAHA.120.030334 PubMed DOI PMC

Santos EM, Yoo AJ, Beenen LF, et al. ; MR CLEAN Investigators. Observer variability of absolute and relative thrombus density measurements in patients with acute ischemic stroke. Neuroradiology 2016;58:133–39 10.1007/s00234-015-1607-4 PubMed DOI PMC

Lalys F, Yan V, Kaladji A, et al. . Generic thrombus segmentation from pre- and post-operative CTA. Int J Comput Assist Radiol Surg 2017;12:1501–10 10.1007/s11548-017-1591-8 PubMed DOI

Lee K, Johnson RK, Yin Y, et al. . Three-dimensional thrombus segmentation in abdominal aortic aneurysms using graph search based on a triangular mesh. Comput Biol Med 2010;40:271–78 10.1016/j.compbiomed.2009.12.002 PubMed DOI PMC

Freiman M, Esses SJ, Joskowicz L, et al. . An iterative model-constrained graph-cut algorithm for abdominal aortic aneurysm thrombus segmentation. In: Proceedings of the 2010 IEEE International Conference on Biomedical Imaging: from Nano to Macro, Rotterdam, the Netherlands. April 14, 2010:672–75

Zhuge F, Rubin GD, Sun S, et al. . An abdominal aortic aneurysm segmentation method: level set with region and statistical information. Med Phys 2006;33:1440–53 10.1118/1.2193247 PubMed DOI

Qazi S, Qazi E, Wilson AT, et al. . Identifying thrombus on non-contrast CT in patients with acute ischemic stroke. Diagnostics (Basel) 2021;11:1919 10.3390/diagnostics11101919 PubMed DOI PMC

Hill MD, Goyal M, Menon BK, et al. ; ESCAPE-NA1 Investigators. Efficacy and Safety of Nerinetide for the Treatment of Acute Ischaemic Stroke (ESCAPE-NA1): a multicentre, double-blind, randomised controlled trial. Lancet 2020;395:878–87 10.1016/S0140-6736(20)30258-0 PubMed DOI

Menon BK, d’Esterre CD, Qazi EM, et al. . Multiphase CT angiography: a new tool for the imaging triage of patients with acute ischemic stroke. Radiology 2015;275:510–20 10.1148/radiol.15142256 PubMed DOI

Ospel JM, Volny O, Qiu W, et al. . Impact of multiphase computed tomography angiography for endovascular treatment decision-making on outcomes in patients with acute ischemic stroke. J Stroke 2021;23:377–87 10.5853/jos.2021.00619 PubMed DOI PMC

Beare R, Lowekamp B, Yaniv Z. Image segmentation, registration and characterization in R with SimpleITK. J Stat Soft 2018;86:8 10.18637/jss.v086.i08 PubMed DOI PMC

Najm M, Kuang H, Federico A, et al. . Automated brain extraction from head CT and CTA images using convex optimization with shape propagation. Comput Methods Programs Biomed 2019;176:1–8 10.1016/j.cmpb.2019.04.030 PubMed DOI

Fedorov A, Beichel R, Kalpathy-Cramer J, et al. . 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging 2012;30:1323–41 10.1016/j.mri.2012.05.001 PubMed DOI PMC

Yushkevich PA, Piven J, Hazlett HC, et al. . User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 2006;31:1116–28 10.1016/j.neuroimage.2006.01.015 PubMed DOI

Warfield SK, Zou KH, Wells WM. Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation. IEEE Trans Med Imaging 2004;23:903–21 10.1109/TMI.2004.828354 PubMed DOI PMC

Ronneberger O, Fischer P, Brox T.. U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab N, Hornegger J, Wells W, et al., eds. In: Medical Image Computing and Computer-Assisted Intervention, Munich, Germany. October 5–9, 2015

Woo S, Park J, Lee JY, et al. . CBAM: Convolutional Block Attention Module. arXiv July 17, 2018. https://arxiv.org/pdf/1807.06521v1.pdf. Accessed October 6, 2018

Feng S, Zhao H, Shi F, et al. . CPFNet: Context Pyramid Fusion Network for Medical Image Segmentation. IEEE Trans Med Imaging 2020;39:3008–18 10.1109/TMI.2020.2983721 PubMed DOI

Ma J, Chen J, Ng M, et al. . Loss odyssey in medical image segmentation. Med Image Anal 2021;71:102035 10.1016/j.media.2021.102035 PubMed DOI

Mourya GK, Gogoi M, Talbar SN, et al. . Cascaded dilated deep residual network for volumetric liver segmentation from CT image. International Journal of E-Health and Medical Communications 2021;12:34–45 10.4018/IJEHMC.2021010103 DOI

Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J 2012;24:69–71 PubMed PMC

Santos EM, Marquering HA, Berkhemer OA, et al. ; MR CLEAN Investigators. Development and validation of intracranial thrombus segmentation on CT angiography in patients with acute ischemic stroke. PLoS One 2014;9:e101985 10.1371/journal.pone.0101985 PubMed DOI PMC

Riedel CH, Jensen U, Rohr A, et al. . Assessment of thrombus in acute middle cerebral artery occlusion using thin-slice nonenhanced computed tomography reconstructions. Stroke 2010;41:1659–64 10.1161/STROKEAHA.110.580662 PubMed DOI

Lucas C, Schöttler JJ, Kemmling A, et al. . Automatic detection and segmentation of the acute vessel thrombus in cerebral CT. Informatik Aktuell Springer; 2019:74–79 10.1007/978-3-658-25326-4_19 DOI

Mojtahedi M, Kappelhof M, Ponomareva E, et al. . Fully automated thrombus segmentation on CT images of patients with acute ischemic stroke. Diagnostics (Basel) 2022;12:698 10.3390/diagnostics12030698 PubMed DOI PMC

Dutra BG, Tolhuisen ML, Alves HC; MR CLEAN Registry Investigators, et al.. Thrombus imaging characteristics and outcomes in acute ischemic stroke patients undergoing endovascular treatment. Stroke 2019;50:2057–64 10.1161/STROKEAHA.118.024247 PubMed DOI

Yoo J, Baek JH, Park H, et al. . Thrombus volume as a predictor of nonrecanalization after intravenous thrombolysis in acute stroke. Stroke 2018;49:2108–15 10.1161/STROKEAHA.118.021864 PubMed DOI

Campbell BC, Kappelhof M, Fischer U. Role of intravenous thrombolytics prior to endovascular thrombectomy. Stroke 2022;53:2085–92 10.1161/STROKEAHA.122.036929 PubMed DOI

Huang C, Tian J, Yuan C, et al. . Fully automated segmentation of lower extremity deep vein thrombosis using convolutional neural network. Biomed Res Int 2019;2019:3401683 10.1155/2019/3401683 PubMed DOI PMC

Santos EM, Niessen WJ, Yoo AJ, et al. ; MR CLEAN Investigators. Automated entire thrombus density measurements for robust and comprehensive thrombus characterization in patients with acute ischemic stroke. PLoS One 2016;11:e0145641 10.1371/journal.pone.0145641 PubMed DOI PMC