YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
🫁 Silicosis Detection System
A 2-stage classification system for detecting and classifying silicosis from chest X-ray images using CXR Foundation embeddings.
Features
- Stage 1: Disease screening (Q2A/Q3A detection)
- Stage 2: Opacity subtyping for disease cases
- Lung region filtering for improved accuracy
- Real-time analysis via Gradio interface
Performance
- Q2A (Parenchymal Disease): 80.7% accuracy
- Q3A (Pleural Abnormalities): 83.6% accuracy
- Opacity Classification: 69.5% ordinal accuracy
Usage
- Upload a chest X-ray image
- Click "Analyze Image"
- Review the 2-stage analysis results
Technical Details
- Uses CXR Foundation model for feature extraction
- Scikit-learn classifiers for disease detection
- Lung region filtering for focused analysis
- 2-stage workflow matching clinical practice
Privacy
This is a private space. Your uploaded images are processed securely and not stored permanently.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support