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: Oct 10, 2025

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

17.0K

Weakly Supervised Sensitive Heatmap framework to classify and localize diabetic retinopathy lesions.

Mohammed Al-Mukhtar1, Ameer Hussein Morad2, Mustafa Albadri1

  • 1Computer Center, University of Baghdad, Baghdad, Iraq.

Scientific Reports
|December 9, 2021
PubMed
Summary

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

FAPNET: Feature Fusion with Adaptive Patch for Flood-Water Detection and Monitoring.

Sensors (Basel, Switzerland)ยท2022
See all related articles

Diabetic retinopathy (DR) detection is improved with a new CNN-WSSH model. This automated system accurately classifies DR and localizes lesions in fundus images, aiding early diagnosis and preventing vision loss.

Area of Science:

  • Ophthalmology
  • Medical Imaging
  • Computer Science

Background:

  • Diabetic retinopathy (DR) is a leading cause of blindness, necessitating early detection.
  • Computer-Aided Diagnosis (CADx) systems are crucial for analyzing fundus images to identify DR features.
  • Timely diagnosis of DR aids medical professionals in making informed treatment decisions.

Purpose of the Study:

  • To develop an automated method for early diabetic retinopathy classification and lesion localization.
  • To combine supervised learning for classification with weakly-supervised learning for localization.
  • To enhance the accuracy and efficiency of DR detection in retinal images.

Main Methods:

  • A four-stage deep learning model was developed, including preprocessing, optic disk segmentation, and DR classification.

More Related Videos

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.8K
Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature
05:54

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature

Published on: January 24, 2025

426

Related Experiment Videos

Last Updated: Oct 10, 2025

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging
06:08

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

17.0K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.8K
Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature
05:54

Quantification of Diabetes-induced Adherent Leukocytes in Retinal Vasculature

Published on: January 24, 2025

426
  • A Convolutional Neural Network (CNN) model was adapted for DR classification using supervised learning.
  • Weakly-supervised learning was employed for lesion localization, incorporating a Weakly Supervised Sensitive Heat Map (WSSH) layer.

    • Main Results:

    • The CNN-WSSH model achieved a test accuracy of 98.65% for lesion detection.
    • The WSSH layer effectively localized the Region of Interest (ROI) of DR lesions.
    • Comparison with Class Activation Map (CAM) showed superior performance for the WSSH approach (0.954).

    Conclusions:

    • The developed CNN-WSSH model demonstrates high accuracy in classifying diabetic retinopathy.
    • The model effectively localizes DR lesions, providing valuable insights for diagnosis.
    • This automated approach aids in early detection and management of diabetic retinopathy, potentially preventing vision loss.