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: Mar 2, 2026

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

3.6K

Deep image mining for diabetic retinopathy screening.

Gwenolé Quellec1, Katia Charrière2, Yassine Boudi2

  • 1Inserm, UMR 1101, 22 avenue Camille-Desmoulins, Brest F-29200, France.

Medical Image Analysis
|May 17, 2017
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

Reply : Evidence-based functional classification of simultaneous vision intraocular lenses: seeking a global consensus by the ESCRS Functional Vision Working Group.

Journal of cataract and refractive surgery·2026
Same author

Correcting Astigmatism Using Toric Intraocular Lenses During Cataract Surgery.

American journal of ophthalmology·2026
Same author

Context-Aware Vision Language Foundation Models for Ocular Disease Screening in Retinal Images.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Acoustic and machine learning methods for speech-based suicide risk assessment: A systematic review.

Journal of affective disorders·2025
Same author

A robust deep learning classifier for screening multiple retinal diseases on optical coherence tomography.

Scientific reports·2025
Same author

Deep learning for retinal non-perfusion and foveal avascular zone analysis in wide-field OCTA in diabetic retinopathy.

Scientific reports·2025
This summary is machine-generated.

This study introduces a new deep learning method to create heatmaps for medical image analysis, improving diabetic retinopathy detection and lesion identification without expert input.

Area of Science:

  • Medical Image Analysis
  • Deep Learning in Healthcare
  • Ophthalmology Imaging

Background:

  • Deep learning, particularly Convolutional Neural Networks (ConvNets), excels in medical image analysis but often functions as a black box.
  • Understanding the specific image regions ConvNets use for diagnosis is crucial for trust and further development.

Purpose of the Study:

  • To develop a method for generating heatmaps that highlight image pixels critical for ConvNet predictions.
  • To enable ConvNets trained for image-level classification to also detect lesions.
  • To improve the interpretability and diagnostic capabilities of deep learning models in medical imaging.

Main Methods:

  • A generalization of the backpropagation method was developed to train ConvNets capable of producing high-quality heatmaps.
Keywords:
Deep learningDiabetic retinopathy screeningImage miningLesion detection

More Related Videos

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

22.4K
Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy
07:45

Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy

Published on: December 30, 2025

544

Related Experiment Videos

Last Updated: Mar 2, 2026

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

3.6K
Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

22.4K
Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy
07:45

Tear-Derived Exosomal miR-15a as New Diagnostic Tool for Diabetic Retinopathy

Published on: December 30, 2025

544
  • The proposed heatmap generation technique was applied to diabetic retinopathy (DR) screening using large datasets (Kaggle, e-ophtha).
  • Performance was evaluated for image-level and lesion-level detection, comparing against existing algorithms.

    • Main Results:

    • The method achieved high detection performance for referable DR (Az=0.954 on Kaggle, Az=0.949 on e-ophtha).
    • The detector outperformed specialized algorithms in detecting four lesion types (microaneurysms, hemorrhages, exudates, cotton-wool spots) on the DiaretDB1 dataset.
    • The proposed detector surpassed existing heatmap generation algorithms at the lesion level.

    Conclusions:

    • The developed deep learning approach effectively generates heatmaps for medical image analysis, enhancing lesion detection capabilities.
    • This method allows ConvNets to function as effective lesion detectors without requiring expert knowledge or manual segmentation.
    • The approach shows promise as an image mining tool for discovering novel biomarkers in medical imaging, integrated into systems like Messidor®.