Skip to main content
Advertisement

Main menu

  • Home
  • Content
    • Current Issue
    • Accepted Manuscripts
    • Article Preview
    • Past Issue Archive
    • Video Articles
    • AJNR Case Collection
    • Case of the Week Archive
    • Case of the Month Archive
    • Classic Case Archive
  • Special Collections
    • AJNR Awards
    • Low-Field MRI
    • Alzheimer Disease
    • ASNR Foundation Special Collection
    • Photon-Counting CT
    • View All
  • Multimedia
    • AJNR Podcasts
    • AJNR SCANtastic
    • Trainee Corner
    • MRI Safety Corner
    • Imaging Protocols
  • For Authors
    • Submit a Manuscript
    • Submit a Video Article
    • Submit an eLetter to the Editor/Response
    • Manuscript Submission Guidelines
    • Statistical Tips
    • Fast Publishing of Accepted Manuscripts
    • Graphical Abstract Preparation
    • Imaging Protocol Submission
    • Author Policies
  • About Us
    • About AJNR
    • Editorial Board
    • Editorial Board Alumni
  • More
    • Become a Reviewer/Academy of Reviewers
    • Subscribers
    • Permissions
    • Alerts
    • Feedback
    • Advertisers
    • ASNR Home

User menu

  • Alerts
  • Log in

Search

  • Advanced search
American Journal of Neuroradiology
American Journal of Neuroradiology

American Journal of Neuroradiology

ASHNR American Society of Functional Neuroradiology ASHNR American Society of Pediatric Neuroradiology ASSR
  • Alerts
  • Log in

Advanced Search

  • Home
  • Content
    • Current Issue
    • Accepted Manuscripts
    • Article Preview
    • Past Issue Archive
    • Video Articles
    • AJNR Case Collection
    • Case of the Week Archive
    • Case of the Month Archive
    • Classic Case Archive
  • Special Collections
    • AJNR Awards
    • Low-Field MRI
    • Alzheimer Disease
    • ASNR Foundation Special Collection
    • Photon-Counting CT
    • View All
  • Multimedia
    • AJNR Podcasts
    • AJNR SCANtastic
    • Trainee Corner
    • MRI Safety Corner
    • Imaging Protocols
  • For Authors
    • Submit a Manuscript
    • Submit a Video Article
    • Submit an eLetter to the Editor/Response
    • Manuscript Submission Guidelines
    • Statistical Tips
    • Fast Publishing of Accepted Manuscripts
    • Graphical Abstract Preparation
    • Imaging Protocol Submission
    • Author Policies
  • About Us
    • About AJNR
    • Editorial Board
    • Editorial Board Alumni
  • More
    • Become a Reviewer/Academy of Reviewers
    • Subscribers
    • Permissions
    • Alerts
    • Feedback
    • Advertisers
    • ASNR Home
  • Follow AJNR on Twitter
  • Visit AJNR on Facebook
  • Follow AJNR on Instagram
  • Join AJNR on LinkedIn
  • RSS Feeds

AJNR Awards, New Junior Editors, and more. Read the latest AJNR updates

Research ArticleAdult Brain

Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke

Y. Yu, Y. Xie, T. Thamm, E. Gong, J. Ouyang, S. Christensen, M.P. Marks, M.G. Lansberg, G.W. Albers and G. Zaharchuk
American Journal of Neuroradiology June 2021, 42 (6) 1030-1037; DOI: https://doi.org/10.3174/ajnr.A7081
Y. Yu
aFrom the Radiology Department (Y.Y., Y.X., T.T., M.P.M., G.Z.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for Y. Yu
Y. Xie
aFrom the Radiology Department (Y.Y., Y.X., T.T., M.P.M., G.Z.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for Y. Xie
T. Thamm
aFrom the Radiology Department (Y.Y., Y.X., T.T., M.P.M., G.Z.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for T. Thamm
E. Gong
bElectrical Engineering Department (E.G., J.O.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for E. Gong
J. Ouyang
bElectrical Engineering Department (E.G., J.O.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for J. Ouyang
S. Christensen
cNeurology Department (S.C., M.G.L., G.W.A.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for S. Christensen
M.P. Marks
aFrom the Radiology Department (Y.Y., Y.X., T.T., M.P.M., G.Z.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for M.P. Marks
M.G. Lansberg
cNeurology Department (S.C., M.G.L., G.W.A.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for M.G. Lansberg
G.W. Albers
cNeurology Department (S.C., M.G.L., G.W.A.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for G.W. Albers
G. Zaharchuk
aFrom the Radiology Department (Y.Y., Y.X., T.T., M.P.M., G.Z.), Stanford University, California
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for G. Zaharchuk
  • Article
  • Figures & Data
  • Supplemental
  • Info & Metrics
  • Responses
  • References
  • PDF
Loading

Article Figures & Data

Figures

  • Tables
  • FIG 1.
    • Download figure
    • Open in new tab
    • Download powerpoint
    FIG 1.

    Flow diagram of the study. According to the reperfusion rate calculated from baseline and 4- to 24- hour perfusion-weighted imaging, patients are grouped into major reperfusion (≥80%), partial reperfusion (20%–80%), minimal reperfusion (≤20%), and unknown reperfusion status (if 4- to 24-hour perfusion imaging was not performed). D1 indicates DEFUSE study; D2, DEFUSE 2 study.

  • FIG 2.
    • Download figure
    • Open in new tab
    • Download powerpoint
    FIG 2.

    Illustration of 3 approaches to define tissue at risk and ischemic core and detailed case distribution during 5-fold cross-validation. The pretraining deep learning approach (A), the separate deep learning approach (B), and the thresholding approach (C). For the pretraining, 94 patients with partial/unknown reperfusion were used in the training set, and 11 patients were in the validation set. D, Cases without 3- to 7-day follow-up were used only in training (A and B), while patients with 3- to 7-day follow-up were used in training, validation, and testing. Patients with 3- to 7-day follow-up were divided randomly into 5 sets and used in training, validation, and testing with a ratio of 3:1:1. For each fold of the tissue-at-risk model, 37 cases were used for fine-tuning on the pretrained model, 6 for validation, and 6–7 for testing. For each fold of the ischemic core model, 48 cases were used for the fine-tuning; 13, for validation; and 13–14, for testing.

  • FIG 3.
    • Download figure
    • Open in new tab
    • Download powerpoint
    FIG 3.

    Two representative cases of predictions from the pretraining, separate, and thresholding approaches. Upper row: A 71 -year-old man treated with IV tPA only, which achieved 0% reperfusion. Of note, only voxels with both diffusion and perfusion information are included in the model. His true lesion is used to define tissue at risk. Lower row: A 45-year-old man treated with IV tPA and thrombectomy, which achieved 100% reperfusion. His true lesion is used to define ischemic core. The pretraining approach had more accurate prediction than either the separate approach or the thresholding method, both visually, with DSC analysis, and volumetrically (Online Supplemental Data). Green areas overlaid on the FLAIR image represent true-positive, blue represents false-negative, and red represents false-positive.

  • FIG 4.
    • Download figure
    • Open in new tab
    • Download powerpoint
    FIG 4.

    Example of how the deep learning models would be used for triage compared with the thresholding method. A, A 44-year-old woman with a large mismatch between tissue at risk and ischemic core prediction (mismatch ratio of 11.8), indicating a small stroke if the patient received successful treatment or a much larger stroke if the patient did not receive any treatment. B, 44-year-old woman with a small mismatch ratio of 1.4, indicating limited tissue salvage despite successful treatment. This illustrates how estimated tissue outcome (with and without reperfusion) can be obtained from the deep learning approach and can facilitate clinical decision-making on whether to treat the patient (see Online Supplemental Data for detailed image).

Tables

  • Figures
  • Clinical and imaging information of patients includeda

    All Patients (n = 237)Minimal Reperfusion (n = 52)Major Reperfusion (n = 80)Partial Reperfusion (n = 57)Unknown Reperfusion (n = 48)P Value
    Male108 (46)27 (52)39 (49)25 (44)17 (35).3
    Age (yr)66 [SD, 16]64 [SD, 16]66 [SD, 16]69 [SD, 14]64 [SD, 17].4
    Hypertension159 (67)39 (75)50 (63)39 (68)31 (65).3
    Diabetes58 (25)14 (27)18 (23)15 (26)11 (23).8
    Dyslipidemia97 (41)24 (46)30 (38)30 (53)13 (28).6
    Atrial fibrillation73 (31)12 (23)27 (34)21 (37)13 (28).4
    Treatment methods.004
        IV tPA only79 (33)25 (48)18 (22)21 (37)15 (31)
        Direct thrombectomy62 (26)8 (15)29 (36)16 (28)9 (19)
        Bridging therapy77 (32)13 (25)31 (39)16 (28)17 (35)
        No treatment19 (8)6 (12)2 (3)4 (7)7 (15)
    Onset-to-treatment time (hr)5.7 (4.7–7.4)5.8 (5.2–6.3)5.8 (4.6–7.8)5.4 (4.6–6.4)5.8 (4.9–7.7).9
    Baseline DWI lesion volume (mL)22 (8–57)20 (6–63)17 (6–43)31 (16–83)31 (13–61).01
    Baseline Tmax lesion volume (mL)115 (68–173)98 (48–158)115 (71–160)126 (66–188)123 (80–171).3
    PWI/DWI mismatch ratiob3.8 (1.9–8.6)2.9 (1.4–6.8)5.4 (2.3–13.9)3.3 (2.0–5.8)3.2 (1.6–6.6).006
    Baseline NIHSS14 (10–19)13 (8–19)15 (9–19)16 (10–19)14 (11–19).3
    Symptomatic hemorrhage27 (11)7 (13)8 (10)8 (14)4 (8).1
    Reperfusion rate (%)69 (15–97)0 (0–9)100 (92–100)55 (37–68)NA<.001
    Final infarct volume (mL)49 (14–108)59 (28–204)19 (8–62)77 (33–149)57 (22–112)<.001
    90-day mRS3 (1–4)3 (2–4)2 (1–3)4 (2–5)3 (1–4)<.001
    • Note:—NA indicates not applicable.

    • ↵a Data are expressed as No. (%), median (IQR), or mean [SD].

    • ↵b The upper limit of the mismatch ratio was set to 20 if a small or no ischemic core lesion presented at baseline.

PreviousNext
Back to top

In this issue

American Journal of Neuroradiology: 42 (6)
American Journal of Neuroradiology
Vol. 42, Issue 6
1 Jun 2021
  • Table of Contents
  • Index by author
  • Complete Issue (PDF)
Advertisement
Print
Download PDF
Email Article

Thank you for your interest in spreading the word on American Journal of Neuroradiology.

NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.

Enter multiple addresses on separate lines or separate them with commas.
Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke
(Your Name) has sent you a message from American Journal of Neuroradiology
(Your Name) thought you would like to see the American Journal of Neuroradiology web site.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Cite this article
Y. Yu, Y. Xie, T. Thamm, E. Gong, J. Ouyang, S. Christensen, M.P. Marks, M.G. Lansberg, G.W. Albers, G. Zaharchuk
Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke
American Journal of Neuroradiology Jun 2021, 42 (6) 1030-1037; DOI: 10.3174/ajnr.A7081

Citation Manager Formats

  • BibTeX
  • Bookends
  • EasyBib
  • EndNote (tagged)
  • EndNote 8 (xml)
  • Medlars
  • Mendeley
  • Papers
  • RefWorks Tagged
  • Ref Manager
  • RIS
  • Zotero
0 Responses
Respond to this article
Share
Bookmark this article
Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke
Y. Yu, Y. Xie, T. Thamm, E. Gong, J. Ouyang, S. Christensen, M.P. Marks, M.G. Lansberg, G.W. Albers, G. Zaharchuk
American Journal of Neuroradiology Jun 2021, 42 (6) 1030-1037; DOI: 10.3174/ajnr.A7081
del.icio.us logo Twitter logo Facebook logo Mendeley logo
  • Tweet Widget
  • Facebook Like
  • Google Plus One
Purchase

Jump to section

  • Article
    • Abstract
    • ABBREVIATIONS:
    • MATERIALS AND METHODS
    • RESULTS
    • DISCUSSION
    • CONCLUSIONS
    • ACKNOWLEDGMENTS
    • Footnotes
    • References
  • Figures & Data
  • Supplemental
  • Info & Metrics
  • Responses
  • References
  • PDF

Related Articles

  • PubMed
  • Google Scholar

Cited By...

  • Deep learning-based detection and segmentation of diffusion abnormalities in acute ischemic stroke
  • Crossref
  • Google Scholar

This article has not yet been cited by articles in journals that are participating in Crossref Cited-by Linking.

More in this TOC Section

Adult Brain

  • Diagnostic Neuroradiology of Monoclonal Antibodies
  • Clinical Outcomes After Chiari I Decompression
  • Segmentation of Brain Metastases with BLAST
Show more Adult Brain

Functional

  • Kurtosis and Epileptogenic Tubers: A Pilot Study
  • Glutaric Aciduria Type 1: DK vs. Conventional MRI
  • Multiparametric MRI in PEDS Pontine Glioma
Show more Functional

Similar Articles

Advertisement

Indexed Content

  • Current Issue
  • Accepted Manuscripts
  • Article Preview
  • Past Issues
  • Editorials
  • Editor's Choice
  • Fellows' Journal Club
  • Letters to the Editor
  • Video Articles

Cases

  • Case Collection
  • Archive - Case of the Week
  • Archive - Case of the Month
  • Archive - Classic Case

More from AJNR

  • Trainee Corner
  • Imaging Protocols
  • MRI Safety Corner
  • Book Reviews

Multimedia

  • AJNR Podcasts
  • AJNR Scantastics

Resources

  • Turnaround Time
  • Submit a Manuscript
  • Submit a Video Article
  • Submit an eLetter to the Editor/Response
  • Manuscript Submission Guidelines
  • Statistical Tips
  • Fast Publishing of Accepted Manuscripts
  • Graphical Abstract Preparation
  • Imaging Protocol Submission
  • Evidence-Based Medicine Level Guide
  • Publishing Checklists
  • Author Policies
  • Become a Reviewer/Academy of Reviewers
  • News and Updates

About Us

  • About AJNR
  • Editorial Board
  • Editorial Board Alumni
  • Alerts
  • Permissions
  • Not an AJNR Subscriber? Join Now
  • Advertise with Us
  • Librarian Resources
  • Feedback
  • Terms and Conditions
  • AJNR Editorial Board Alumni

American Society of Neuroradiology

  • Not an ASNR Member? Join Now

© 2025 by the American Society of Neuroradiology All rights, including for text and data mining, AI training, and similar technologies, are reserved.
Print ISSN: 0195-6108 Online ISSN: 1936-959X

Powered by HighWire