Downloads · 30 days
0
cataract-resnet18-classifier
amanchandra/cataract-resnet18-classifier
cataract-resnet18-classifier is a image classification model from amanchandra. Use it when you need a label for an image. The card lists the license as mit.
Trained on ~1170 clinical images (binary: cataract vs non-cataract).
Access
Public
Updated Jun 27, 2025
Repo size
44.8 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth44.8 MB · 100%
At a glance
- Task
- Image Classification
- License
- mit
- Access
- Public
- Created
- Jun 27, 2025
- Updated
- Jun 27, 2025
- SHA
- c3252158
Datasets
Model card
From the Hugging Face model README
Cataract Detection ResNet18 Model
Trained on ~1170 clinical images (binary: cataract vs non-cataract).
Architecture: ResNet18
Image Size: 224x224
Train Accuracy: 100%
Test Accuracy: 97.16%
Use: Baseline model for severity grading project
Future Steps
- Grad-CAM to interpret focus region (pupil/lens)
- Region-based unsupervised feature extraction
- Clustering for cataract severity grading
- Task
- Image Classification
- License
- mit
- Languages
- en
- Created
- Jun 27, 2025
- Updated
- Jun 27, 2025