Datasets:
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Error code: JobManagerCrashedError
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
image
image |
|---|
EyePACS β Diabetic Retinopathy Fundus Image Dataset
Image: EYEPACS Preprocess Samples. |
π Overview
EyePACS (Eye Picture Archive Communication System) is a large-scale collection of retinal fundus images used for automated diabetic retinopathy (DR) detection.
It formed the basis of the Kaggle Diabetic Retinopathy Detection challenge, enabling research into DR classification and screening.
The dataset includes macula-centered color fundus images from real-world clinical screenings under varied imaging conditions.
π Dataset Summary
| Field | Details |
|---|---|
| Task | Diabetic retinopathy classification (5 severity grades) |
| Description | High-resolution color fundus photographs from EyePACS DR screening program. Each image labeled by ophthalmologists using the ICDR scale. |
| Size | ~88,702 images total (β35k labeled for training, β53k unlabeled for testing) |
| Classes | 0 = No DR, 1 = Mild, 2 = Moderate, 3 = Severe, 4 = Proliferative DR |
| Image Type | Macula-centered color fundus photos (640Γ480 to 5184Γ3456 px) |
| Source | EyePACS (USA) via Kaggle Diabetic Retinopathy Detection Challenge (2015) |
| Access | Kaggle Dataset |
| License | Usage restricted under Kaggle & EyePACS terms |
π§± Dataset Structure
eyepacs/
βββ images/
β βββ train/
β β βββ 00001_left.jpg
β β βββ 00001_right.jpg
β βββ test/
βββ labels.csv # image_id, eye(L/R), dr_grade (0β4)
βββ README.md
βββ LICENSE.txt
π§© Label Details
- Labels follow the International Clinical Diabetic Retinopathy (ICDR) grading system.
- Class imbalance is significant β most images show no DR.
- Labels were assigned by certified ophthalmologists; minor label noise may exist.
βοΈ Preprocessing Recommendations
- Crop to the circular fundus region and remove borders
- Resize to consistent resolution (e.g. 1024Γ1024)
- Normalize illumination and contrast
- Exclude blurred or ungradable images
π‘ Research Applications
- DR detection and severity classification
- Automated retinal screening systems
- Transfer learning and robustness testing across imaging conditions
- Comparative studies with datasets like MESSIDOR, DDR, and APTOS
β οΈ Notes & Limitations
- Significant class imbalance (majority = No DR)
- Variations in camera type, exposure, and focus
- Some label noise and ungradable images present
- Redistribution may be restricted β verify Kaggle/EyePACS terms before publishing images
π Citation
If you use the dataset, cite:
Kaggle and EyePACS. βDiabetic Retinopathy Detection.β Kaggle Competition, 2015.
https://www.kaggle.com/c/diabetic-retinopathy-detection
π References
- EyePACS Official Data Page β https://www.eyepacs.com/data-analysis
- Kaggle: Diabetic Retinopathy Detection β https://www.kaggle.com/c/diabetic-retinopathy-detection
- Research overview: Transfer Learning Based Classification of Diabetic Retinopathy on the Kaggle EyePACS Dataset, ResearchGate, 2021.
- Downloads last month
- 69