1. Input image validation (fundus / non-fundus)
The model first checks whether the uploaded image is a valid retinal fundus photograph.
Fundus Image Analysis – Research Demo
Prototype tool for automatic segmentation of the optic cup and optic disc in retinal fundus images. Designed solely for research and educational purposes.
Disclaimer: This web demo is not a medical device and does not provide medical diagnosis, screening, or treatment advice. Results are research outputs only and must not be used as the sole basis for any clinical or health-related decision. Users should always consult a licensed eye-care professional for medical advice.
2. Optic disc localisation – research model
The research model highlights a candidate region for the optic disc in the original image.
3. Region of interest around optic disc (×1.5 crop)
A cropped region centred on the predicted optic disc, used as input for segmentation.
4. Resized with padding for segmentation model
The region of interest is resized and padded to match the expected input size of the model.
5. Segmentation result – research output only
Experimental segmentation of the optic cup and disc. Outputs are not validated for
clinical decision-making.
Uploaded images are processed transiently on the research server of the Institute of Information and Computational Technologies. Images are not stored for user-tracking or commercial purposes.
For more details, please see our Privacy & Research Use Policy.