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: Sep 10, 2025

Author Spotlight: Insights into Visual Cortex Research Through Wide-View fMRI Mapping
07:11

Author Spotlight: Insights into Visual Cortex Research Through Wide-View fMRI Mapping

Published on: December 8, 2023

1.9K

Multi-stage framework using transformer models, feature fusion and ensemble learning for enhancing eye disease

Abdulaziz AlMohimeed1

  • 1College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi Arabia. aialmohimeed@imamu.edu.sa.

Scientific Reports
|August 19, 2025
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

Cloud-based real-time enhancement for disease prediction using Confluent Cloud, Apache Kafka, feature optimization, and explainable artificial intelligence.

PeerJ. Computer science·2025
Same author

Advancing arabic dialect detection with hybrid stacked transformer models.

Frontiers in human neuroscience·2025
Same author

ViT-PSO-SVM: Cervical Cancer Predication Based on Integrating Vision Transformer with Particle Swarm Optimization and Support Vector Machine.

Bioengineering (Basel, Switzerland)·2024
Same author

Automated Diagnosis for Colon Cancer Diseases Using Stacking Transformer Models and Explainable Artificial Intelligence.

Diagnostics (Basel, Switzerland)·2023
Same author

Diagnosis of COVID-19 Using Chest X-ray Images and Disease Symptoms Based on Stacking Ensemble Deep Learning.

Diagnostics (Basel, Switzerland)·2023

This study introduces a multi-stage deep learning framework (MST-EDS) for accurate eye disease diagnosis. The proposed model achieved 97.163% accuracy in classifying normal, diabetic retinopathy, glaucoma, and cataract images.

Area of Science:

  • Ophthalmology
  • Computer Science
  • Artificial Intelligence

Background:

  • Early diagnosis of eye diseases is vital for preventing vision impairment.
  • Deep learning models show potential for automated eye disease diagnosis from images.
  • Current single-model architectures may struggle with complex feature extraction for accurate classification.

Purpose of the Study:

  • To propose a novel multi-stage framework (MST-EDS) for improved eye disease classification.
  • To address limitations of single-model architectures in capturing spatial and fine-grained features.
  • To categorize eye illnesses into normal, diabetic retinopathy, glaucoma, and cataract.

Main Methods:

  • Developed a two-stage framework: hybrid models and stacking models.
Keywords:
Data-efficient image transformer (DeiT)Diagnostic modelEye diseasesImage processingMST-EDSSwin transformerVision transformer (ViT)

More Related Videos

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.0K

Related Experiment Videos

Last Updated: Sep 10, 2025

Author Spotlight: Insights into Visual Cortex Research Through Wide-View fMRI Mapping
07:11

Author Spotlight: Insights into Visual Cortex Research Through Wide-View fMRI Mapping

Published on: December 8, 2023

1.9K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.0K
  • Hybrid models utilized Transformer architectures (ViT, DeiT, Swin) for feature extraction, PCA for dimensionality reduction, and ML classifiers.
  • Stacking models combined outputs from best hybrid models to train meta-learners for enhanced performance.

    • Main Results:

    • The MST-EDS framework, specifically the MST-EDS-RF model, outperformed individual Transformer and hybrid models.
    • Achieved a classification accuracy of 97.163% on a benchmark Kaggle dataset.
    • Demonstrated the effectiveness of multi-stage and stacking approaches in deep learning for medical imaging.

    Conclusions:

    • The proposed MST-EDS framework offers a robust and accurate solution for automated eye disease diagnosis.
    • Multi-stage and stacking ensemble methods significantly enhance classification performance compared to single models.
    • This approach holds promise for improving early detection and management of various eye conditions.