Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Dec 30, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

Tournament Based Ranking CNN for the Cataract grading.

Dohyeun Kim, Tae Joon Jun, Youngsub Eom

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |January 18, 2020
    PubMed
    Summary

    A new Tournament based Ranking CNN significantly improves medical image classification accuracy for underrepresented classes. This deep learning model enhances identification of severe disease cases, outperforming existing methods.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Modulatory Effects of Procyanidin B1 on Inflammation-Induced Oxidative Stress and ECM-Related Responses in Human Dermal Fibroblasts and Epidermal Keratinocytes.

    Molecules (Basel, Switzerland)·2026
    Same author

    Deep learning-based treatment decision support framework for multi-vessel coronary artery disease using integrated coronary angiography and clinical data.

    BMC medical informatics and decision making·2026
    Same author

    Pulmonary carcinogenic effects of subchronic inhalation exposure with long-term follow-up to polyhexamethylene guanidine phosphate in rats.

    Scientific reports·2026
    Same author

    Evaluation of anemia in non-enhanced and contrast-enhanced dual-energy CT using electron density imaging.

    PloS one·2026
    Same author

    Iodinated Contrast Media Hypersensitivity in 115,966 Patients: Risk Factors, Severity Profiles, and the Impact of Iodine Concentration on Reaction Risk.

    Investigative radiology·2026
    Same author

    Chronic neuroinflammation after acute SARS-Cov-2 infection induces retinal damage in the hACE2 transgenic mouse model.

    Journal of neuroimmunology·2026

    Area of Science:

    • Computer Science
    • Medical Imaging
    • Artificial Intelligence

    Background:

    • Class imbalance in medical datasets leads to poor performance in identifying underrepresented categories, particularly severe disease cases.
    • Vagueness and overlapping features between medical data grades complicate precise classification.
    • Existing methods like Ranking CNN struggle with medical data due to aggregation issues.

    Purpose of the Study:

    • To propose a novel Convolutional Neural Network (CNN) architecture, Tournament based Ranking CNN, to address performance degradation caused by class imbalance in medical datasets.
    • To enhance the accurate classification of underrepresented or severe disease classes.
    • To improve upon the Ranking CNN method for medical data applications.

    Main Methods:

    Related Experiment Videos

    Last Updated: Dec 30, 2025

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

    Published on: November 30, 2022

    3.3K
  • Developed a new CNN architecture named Tournament based Ranking CNN.
  • Implemented a tournament structure for aggregating outputs from multiple deep, pre-trained binary neural network models.
  • Applied the model to cataract grading, a medical problem with ordinal labels and imbalanced data.

    • Main Results:

    • The proposed Tournament based Ranking CNN achieved 68.36% exact match accuracy.
    • This significantly outperformed Ranking CNN (53.40%), pre-trained ResNet (56.12%), and CNN with linear regression (57.48%).
    • The model demonstrated a remarkable performance gain in identifying dominated classes with minimal accuracy loss in dominating classes.

    Conclusions:

    • The Tournament based Ranking CNN effectively addresses class imbalance and vagueness in medical data classification.
    • The proposed method shows high efficiency in cataract grading and is adaptable to other medical problems with similar data characteristics.
    • This deep learning approach offers a promising solution for improving diagnostic accuracy in challenging medical imaging scenarios.