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 ArticleHead and Neck Imaging
Open Access

Convolutional Neural Network to Stratify the Malignancy Risk of Thyroid Nodules: Diagnostic Performance Compared with the American College of Radiology Thyroid Imaging Reporting and Data System Implemented by Experienced Radiologists

G.R. Kim, E. Lee, H.R. Kim, J.H. Yoon, V.Y. Park and J.Y. Kwak
American Journal of Neuroradiology May 2021, DOI: https://doi.org/10.3174/ajnr.A7149
G.R. Kim
aFrom the Department of Radiology (G.R.K., J.H.Y., V.Y.P., J.Y.K.), Severance Hospital, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for G.R. Kim
E. Lee
bDepartment of Computational Science and Engineering (E.L.), Yonsei University, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for E. Lee
H.R. Kim
cBiostatistics Collaboration Unit (H.R.K.), Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for H.R. Kim
J.H. Yoon
aFrom the Department of Radiology (G.R.K., J.H.Y., V.Y.P., J.Y.K.), Severance Hospital, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for J.H. Yoon
V.Y. Park
aFrom the Department of Radiology (G.R.K., J.H.Y., V.Y.P., J.Y.K.), Severance Hospital, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for V.Y. Park
J.Y. Kwak
aFrom the Department of Radiology (G.R.K., J.H.Y., V.Y.P., J.Y.K.), Severance Hospital, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Korea
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for J.Y. Kwak
  • Article
  • Figures & Data
  • Info & Metrics
  • Responses
  • References
  • PDF
Loading

Article Figures & Data

Figures

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

    Flowchart of the study population. AUS/FLUS indicates atypia of undetermined significance/follicular lesion of undetermined significance.

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

    Comparison of ROC curves between the CNN (solid line) and ACR TI-RADS categories (dotted line). The area under the ROC curve of the CNN (0.917; 95% confidence interval, 0.895–0.936) was higher than that in the ACR TI-RADS categories (0.891; 95% confidence interval, 0.867–0.912) (P = .017). The areas under the ROC curve of the CNN using a malignancy risk percentage between 0 and 100 and ACR TI-RADS categories using a TR category from 1 to 5 were compared as continuous values.

Tables

  • Figures
    • View popup
    Table 1:

    Patient demographics and distribution of ACR TI-RADS features in benign and malignant thyroid nodules (n = 760)a

    CharacteristicsAll (n = 760)Benign Nodules (n = 584)Malignant Nodules (n = 176)P Value
    Sex.035
        Women587462 (79.4%)125 (71.4%)
        Men170120 (20.6%)50 (28.6%)
    Age (median) (interquartile range) (yr)51 (39–61)52 (41–61)45 (34–60)<.001
    Nodule size (median) (interquartile range) (mm)20 (14–30)23 (15–32)14 (11–20)<.001
    Nodule features
        Composition<.001
            Cystic or almost completely cystic5047 (8.0%)3 (1.7%)
            Spongiform11 (0.2%)0
            Mixed cystic and solid234222 (38.0%)12 (6.8%)
            Solid or almost completely solid475314 (53.8%)161 (91.5%)
        Echogenicity<.001
            Anechoic00
            Hyperechoic or isoechoic410387 (66.3%)23 (13.1%)
            Hypoechoic329191 (32.7%)138 (78.4%)
            Very hypoechoic216 (1.0%)15 (8.5%)
        Shape<.001
            Wider-than-tall671554 (94.9%)117 (66.5%)
            Taller-than-wide8930 (5.1%)59 (33.5%)
        Margin<.001
            Smooth579535 (91.6%)44 (25.0%)
            Ill-defined000
            Lobulated or irregular18149 (8.4%)132 (75.0%)
            Extrathyroidal extension000
        Echogenic foci<.001
            None or large comet-tail artifacts536477 (81.7%)59 (33.5%)
            Macrocalcifications9169 (11.8%)22 (12.5%)
            Peripheral (rim) calcifications1010 (1.7%)0
            Punctate echogenic foci12328 (4.8%)95 (54.0%)
    • ↵a Data are numbers of nodules, with percentages in parentheses.

    • View popup
    Table 2:

    Calculated malignancy risk of each category according to the risk stratification of ACR TI-RADS

    TR1TR2TR3TR4TR5Total
    Suggested risk of malignancy (%)20,24≤2≤22< and ≤55< and ≤20>20
    ACR TI-RADS category
        No. of malignant nodules04933130176
        Assigned total nodules41158185209167760
        Calculated risk of malignancy (%)02.54.915.877.823.2
    CNNa
        No. of malignant nodules0009167176
        Assigned total nodules0545307403760
        Calculated risk of malignancy (%)0002.941.423.2
    • ↵a Malignancy percentages provided by the CNN were re-categorized according to the suggested cancer risk levels of ACR TI-RADS.

    • View popup
    Table 3:

    Comparison of diagnostic performance between CNN and ACR TI-RADS

    CNN (95% CI)ACR TI-RADS (95% CI)P Value
    Sensitivity81.8% (76.1–87.5)73.9% (67.4–80.4).009
    Specificity86.1% (83.3–88.9)93.7% (91.7–95.6)<.001
    Accuracy85.1% (82.6–87.7)89.1% (86.9–91.3).003
    Positive predictive value64.0% (57.7–70.3)77.8% (71.6–84.1)<.001
    Negative predictive value94.0% (92–96)92.2% (90.1–94.4).046
    AUCa0.917 (0.895-0.936)0.891 (0.867-0.912).017
    • ↵a The AUCs of the CNN using a malignancy risk percentage between 0 and 100 and ACR TI-RADS categories using a TR category from 1 to 5 were compared as continuous values.

PreviousNext
Back to top
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.
Convolutional Neural Network to Stratify the Malignancy Risk of Thyroid Nodules: Diagnostic Performance Compared with the American College of Radiology Thyroid Imaging Reporting and Data System Implemented by Experienced Radiologists
(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
G.R. Kim, E. Lee, H.R. Kim, J.H. Yoon, V.Y. Park, J.Y. Kwak
Convolutional Neural Network to Stratify the Malignancy Risk of Thyroid Nodules: Diagnostic Performance Compared with the American College of Radiology Thyroid Imaging Reporting and Data System Implemented by Experienced Radiologists
American Journal of Neuroradiology May 2021, DOI: 10.3174/ajnr.A7149

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
Convolutional Neural Network to Stratify the Malignancy Risk of Thyroid Nodules: Diagnostic Performance Compared with the American College of Radiology Thyroid Imaging Reporting and Data System Implemented by Experienced Radiologists
G.R. Kim, E. Lee, H.R. Kim, J.H. Yoon, V.Y. Park, J.Y. Kwak
American Journal of Neuroradiology May 2021, DOI: 10.3174/ajnr.A7149
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
    • Footnotes
    • References
  • Figures & Data
  • Info & Metrics
  • Responses
  • References
  • PDF

Related Articles

  • No related articles found.
  • PubMed
  • Google Scholar

Cited By...

  • No citing articles found.
  • Crossref (12)
  • Google Scholar

This article has been cited by the following articles in journals that are participating in Crossref Cited-by Linking.

  • AI in Thyroid Cancer Diagnosis: Techniques, Trends, and Future Directions
    Yassine Habchi, Yassine Himeur, Hamza Kheddar, Abdelkrim Boukabou, Shadi Atalla, Ammar Chouchane, Abdelmalik Ouamane, Wathiq Mansoor
    Systems 2023 11 10
  • Artificial Intelligence in Thyroid Field—A Comprehensive Review
    Fabiano Bini, Andrada Pica, Laura Azzimonti, Alessandro Giusti, Lorenzo Ruinelli, Franco Marinozzi, Pierpaolo Trimboli
    Cancers 2021 13 19
  • Accuracy of Ultrasound Diagnosis of Thyroid Nodules Based on Artificial Intelligence-Assisted Diagnostic Technology: A Systematic Review and Meta-Analysis
    Yu Xue, Ying Zhou, Tingrui Wang, Huijuan Chen, Lingling Wu, Huayun Ling, Hong Wang, Lijuan Qiu, Dongqing Ye, Bin Wang, Claudio Casella
    International Journal of Endocrinology 2022 2022
  • Deep Learning and Minimally Invasive Endoscopy: Automatic Classification of Pleomorphic Gastric Lesions in Capsule Endoscopy
    Miguel Mascarenhas, Francisco Mendes, Tiago Ribeiro, João Afonso, Pedro Cardoso, Miguel Martins, Hélder Cardoso, Patrícia Andrade, João Ferreira, Miguel Mascarenhas Saraiva, Guilherme Macedo
    Clinical and Translational Gastroenterology 2023 14 10
  • Diagnosing thyroid nodules with atypia of undetermined significance/follicular lesion of undetermined significance cytology with the deep convolutional neural network
    Inyoung Youn, Eunjung Lee, Jung Hyun Yoon, Hye Sun Lee, Mi-Ri Kwon, Juhee Moon, Sunyoung Kang, Seul Ki Kwon, Kyong Yeun Jung, Young Joo Park, Do Joon Park, Sun Wook Cho, Jin Young Kwak
    Scientific Reports 2021 11 1
  • Diagnostic Performance of ACR and Kwak TI-RADS for Benign and Malignant Thyroid Nodules: An Update Systematic Review and Meta-Analysis
    Yun Jin Kang, Gulnaz Stybayeya, Ju Eun Lee, Se Hwan Hwang
    Cancers 2022 14 23
  • Application of Artificial Intelligence to Evaluate Thyroid Nodules
    Jungmin Ahn, Myung-Chul Lee
    Journal of Clinical Otolaryngology Head and Neck Surgery 2023 34 2
  • Diagnostic efficiency among Eu-/C-/ACR-TIRADS and S-Detect for thyroid nodules: a systematic review and network meta-analysis
    Longtao Yang, Cong Li, Zhe Chen, Shaqi He, Zhiyuan Wang, Jun Liu
    Frontiers in Endocrinology 2023 14
  • Graphical user interface-based convolutional neural network models for detecting nasopalatine duct cysts using panoramic radiography
    Kotaro Ito, Naohisa Hirahara, Hirotaka Muraoka, Eri Sawada, Satoshi Tokunaga, Tomohiro Komatsu, Takashi Kaneda
    Scientific Reports 2024 14 1
  • Personalized Diagnosis in Differentiated Thyroid Cancers by Molecular and Functional Imaging Biomarkers: Present and Future
    Laura Teodoriu, Letitia Leustean, Maria-Christina Ungureanu, Stefana Bilha, Irena Grierosu, Mioara Matei, Cristina Preda, Cipriana Stefanescu
    Diagnostics 2022 12 4

More in this TOC Section

  • ASL Sensitivity for Head and Neck Paraganglioma
  • Post SRS Peritumoral Hyperintense Signal of VSs
  • Normal Facial Nerve Enhancement on VIBE Sequence
Show more Head and Neck Imaging

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