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: Jun 10, 2026

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

Expert-Guided Cross-View Fusion With Self-Derived Lesion Proposals for Multi-View Diabetic Retinopathy Grading.

Wai Keung Wong, Wenzhe Liu, Xueling Zhou

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |June 8, 2026
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    Diabetic Retinopathy01:27

    Diabetic Retinopathy

    DefinitionDiabetic retinopathy is a microvascular complication of diabetes affecting the retinal blood vessels.Risk FactorsDiabetic retinopathy is present in almost all individuals with type 1 diabetes and more than 60% of those with type 2 diabetes after two decades of disease.The risk increases with poor glycemic control, hypertension, dyslipidemia, smoking, pregnancy, and puberty.Although cataracts and glaucoma are also more frequent in people with diabetes, retinopathy remains the leading...

    You might also read

    Related Articles

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

    Sort by
    Same author

    Therapeutic Effects of Combined Treatment with the AEA Hydrolysis Inhibitor PF04457845 and the Substrate Selective COX-2 Inhibitor LM4131 in the Mouse Model of Neuropathic Pain.

    Cells·2023
    Same author

    Effects of different pelvic osteotomies on acetabular morphology in developmental dysplasia of hip in children.

    World journal of orthopedics·2023
    Same author

    Disheveled3 enhanced EMT and cancer stem-like cells properties via Wnt/β-catenin/c-Myc/SOX2 pathway in colorectal cancer.

    Journal of translational medicine·2023
    Same author

    Construction of Curcumin and Paclitaxel Co-Loaded Lipid Nano Platform and Evaluation of Its Anti-Hepatoma Activity in vitro and Pharmacokinetics in vivo.

    International journal of nanomedicine·2023
    Same author

    Identification of an EMT-related gene-based prognostic signature in osteosarcoma.

    Cancer medicine·2023
    Same author

    Community composition and abundance of complete ammonia oxidation (comammox) bacteria in the Lancang River cascade reservoir.

    Ecotoxicology and environmental safety·2023

    This study introduces a new AI framework for diabetic retinopathy (DR) grading that generates its own lesion proposals, reducing the need for costly manual annotations and improving grading accuracy.

    Area of Science:

    • Ophthalmology
    • Medical Imaging
    • Artificial Intelligence

    Background:

    • Multi-view fundus imaging shows promise for automated diabetic retinopathy (DR) grading.
    • Current deep learning models struggle with small, low-contrast lesions due to spatial detail compression.
    • External annotations for lesions or vessels improve performance but are expensive and reduce model independence.

    Purpose of the Study:

    • To develop an end-to-end framework for DR grading that reduces reliance on external annotations.
    • To enable automated lesion proposal generation within the grading pipeline.
    • To enhance the performance of automated DR grading systems by incorporating self-derived cues.

    Main Methods:

    • Proposed a Grade-Activated Lesion Proposal (GALP) module to generate grade-conditioned evidence maps (GEMs) and select top-K lesion proposals.

    Related Experiment Videos

    Last Updated: Jun 10, 2026

    Using Retinal Imaging to Study Dementia
    09:17

    Using Retinal Imaging to Study Dementia

    Published on: November 6, 2017

  • Introduced a Cross-View Lesion Expert Guided Regional Fusion (LGRF) module for selective expert activation based on inter-view guidance.
  • Developed an end-to-end framework integrating GALP and LGRF for on-the-fly lesion proposal generation during training and inference.

    • Main Results:

    • The proposed method achieved performance matching or surpassing strong baselines on two multi-view DR datasets.
    • The framework successfully reduced the need for expensive external annotations.
    • Self-generated lesion proposals significantly contributed to improved DR grading accuracy.

    Conclusions:

    • Automated generation of lesion proposals can substantially decrease annotation requirements in DR grading.
    • The proposed end-to-end framework offers a more efficient and independent approach to automated DR grading.
    • This method holds potential for wider adoption of AI in diabetic retinopathy screening.