Skip to content

Dataset · ctmedtech

Fundus images for retinopathy diagnosis

ctmedtech/DDR-dataset

Medical retina images annotated with diabetic retinopathy stages and pathologies.

What's inside

The DDR dataset contains between 10k and 100k fundus images with English descriptions. Data is provided with both image-level disease severity labels and pixel-level segmentation of individual pathological areas.

What you can do with it

  • Training classification models for diabetic retinopathy stages
  • Segmentation and detection of small retinal lesions in medical images
Tasks
Image segmentation, Image classification, Object detection
Languages
English
Size
10K–100K rows
Downloads this month
28.3K
Likes
2

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("ctmedtech/DDR-dataset")