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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Optimizing compressive strength prediction using adversarial learning and hybrid regularization.

Scientific reports·2024
Same author

Comparing Conventional and Deep Feature Models for Classifying Fundus Photography of Hemorrhages.

Journal of healthcare engineering·2022
Same author

A False Alarm Reduction Method for a Gas Sensor Based Electronic Nose.

Sensors (Basel, Switzerland)·2017
See all related articles

Related Experiment Video

Updated: Aug 12, 2025

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.7K

Deep learning-based hemorrhage detection for diabetic retinopathy screening.

Tamoor Aziz1, Chalie Charoenlarpnopparut2, Srijidtra Mahapakulchai3

  • 1Sirindhorn International Institute of Technology, Thammasat University, Bangkok, Thailand. tamoor.azi@dome.tu.ac.th.

Scientific Reports
|January 27, 2023
PubMed
Summary
This summary is machine-generated.

Early detection of diabetic retinopathy (DR) is crucial. A novel deep learning method accurately identifies hemorrhages, aiding ophthalmologists and potentially preventing vision loss.

More Related Videos

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

2.9K
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

369

Related Experiment Videos

Last Updated: Aug 12, 2025

Using Retinal Imaging to Study Dementia
09:17

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

21.7K
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

2.9K
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

369

Area of Science:

  • Ophthalmology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Diabetic retinopathy (DR) is a leading cause of visual impairment.
  • Hemorrhage is a key pathological symptom of DR, indicating disease presence.
  • Early diagnosis of DR is vital for timely treatment and preventing vision loss.

Purpose of the Study:

  • To propose an automatic deep learning-based method for hemorrhage detection in fundus images.
  • To enhance image quality and identify potential hemorrhage locations.
  • To serve as a supplementary tool for ophthalmologists, reducing screening time and complexity.

Main Methods:

  • Image preprocessing included modified gamma correction for adaptive illumination and gradient information.
  • Potential hemorrhage locations were estimated using Gaussian match filter, entropy thresholding, and mathematical morphology.
  • A novel hemorrhage network was developed for classification and compared against established deep learning models (LeNet-5, AlexNet, ResNet50, VGG-16).

Main Results:

  • The proposed shallow network achieved competitive performance against deeper, renowned models.
  • The method demonstrated promising classification accuracy and significantly reduced training time.
  • Evaluation metrics included sensitivity, specificity, precision, and accuracy on two benchmark datasets.

Conclusions:

  • Increasing deep network layers does not necessarily improve results and can increase training time.
  • The study highlights the critical role of suitable deep model architecture and parameters for optimal outcomes.
  • The developed deep learning approach offers an efficient and accurate method for early diabetic retinopathy hemorrhage detection.